chore(client): regenerate fork types with merged generator fix

packages/ai/AGENTS.md

@@ -165,10 +165,6 @@ Native chronological system messages are route/model-specific. Open Responses lo
 
 The wrapped-user fallback preserves ordering while visibly lowering authority. Never silently pass a raw chronological `role: "system"` through a route that might reject it. Do not insert raw retrieved documents, tool output, or web content into privileged chronological system updates; keep untrusted content in ordinary user/tool channels.
 
-### Effort Updates
-
-`Message.effort({ effort, previous })` is a chronological "reasoning effort changed here" marker (`undefined` means the model default). Changing a top-level effort invalidates the whole provider prompt cache, so protocols with a native per-message update (`Protocol.supportsEffortUpdates`) keep the top-level effort at the first marker's `previous` and lower each marker in place: Anthropic Messages emits an empty `role: "system"` message with `output_config.effort` plus the `mid-conversation-output-config-2026-07-01` beta, and OpenAI Responses emits `configuration_update` items. `applyEffortUpdates` runs in `prepareRequest` and strips the markers for every other route, so a protocol without support keeps today's plain top-level behaviour. When the last marker disagrees with the effort the request asks for (reverted or forked history), `resolveEffortUpdates` strips the markers and falls back to a plain top-level change.
-
 ### Tools
 
 Tool loops are represented in common messages and events:

packages/ai/README.md

@@ -29,78 +29,6 @@ await Effect.runPromise(program.pipe(Effect.provide(llmLayer)))
 
 Run `LLMClient.stream(request)` instead of `generate` when you want incremental `LLMEvent`s. The event stream is provider-neutral — same shape across OpenAI Chat, OpenAI Responses, Anthropic Messages, Gemini, Bedrock Converse, and any OpenAI-compatible deployment.
 
-## Alibaba Cloud Model Studio
-
-`Alibaba` provides standard Model Studio inference. Configure a region explicitly, then select
-Chat Completions (`.model` or `.chat`), Anthropic-compatible Messages (`.messages`), or OpenAI-compatible
-Responses (`.responses`). These routes use HTTP/SSE.
-
-```ts
-import { LLM } from "@opencode/ai"
-import { Alibaba } from "@opencode/ai/providers"
-
-const alibaba = Alibaba.configure({
-  region: "ap-southeast-1", // Singapore
-  apiKey: process.env.DASHSCOPE_API_KEY,
-  // workspaceID: "llm-your-workspace", // use a workspace-dedicated endpoint
-})
-
-const request = LLM.request({
-  model: alibaba.model("qwen3.8-max"),
-  prompt: "Explain this design.",
-  providerOptions: { reasoningEffort: "medium" },
-})
-```
-
-### Regions and credentials
-
-| Region              | `region`         | Shared host when `workspaceID` is omitted |
-| ------------------- | ---------------- | ----------------------------------------- |
-| Singapore           | `ap-southeast-1` | `dashscope-intl.aliyuncs.com`             |
-| China (Beijing)     | `cn-beijing`     | `dashscope.aliyuncs.com`                  |
-| China (Hong Kong)   | `cn-hongkong`    | `cn-hongkong.dashscope.aliyuncs.com`      |
-| US (Virginia)       | `us-east-1`      | `dashscope-us.aliyuncs.com`               |
-| Germany (Frankfurt) | `eu-central-1`   | Supply `workspaceID` or `baseURL`         |
-| Japan (Tokyo)       | `ap-northeast-1` | Supply `workspaceID` or `baseURL`         |
-
-With `workspaceID`, the host is `{workspaceID}.{region}.maas.aliyuncs.com`. A complete `baseURL`
-overrides regional setup, including the API prefix: `/compatible-mode/v1` for Chat/Responses,
-or `/apps/anthropic/v1` for Messages. The selector appends its operation path.
-
-Keys and model availability are region-specific. Auth resolves from explicit `auth` or `apiKey`,
-then `DASHSCOPE_API_KEY`, then `ALIBABA_API_KEY`.
-
-The access region and inference scope differ: Virginia's `-us` model IDs request US-only inference;
-some regions select scope through their workspace. Model IDs pass through unchanged.
-Alibaba's [regional guide](https://www.alibabacloud.com/help/en/model-studio/regions) and
-[base URL table](https://www.alibabacloud.com/help/en/model-studio/base-url) disagree about Virginia's
-shared host; the entry above follows the base URL table. Dedicated hosts can be copied from the console.
-
-### Native options
-
-- **Chat:** `reasoningEffort` → `reasoning_effort`, `enableThinking` → `enable_thinking`,
-  `thinkingBudget` → `thinking_budget`, and `preserveThinking` → `preserve_thinking`.
-  Replay complete `response.message` values to retain `reasoning_content` separately from answer text.
-  Qwen 3.8 defaults to preserving thinking; older models have different defaults.
-  Additional options include `toolStream`, `parallelToolCalls`, `repetitionPenalty`, `responseFormat`,
-  `enableSearch`, and native `searchOptions`. `generation.topK` lowers to `top_k`.
-  `clearThinking` is a hosted GLM control, and `thinking.type` is available for hosted MiniMax models.
-- **Messages:** `effort` → `output_config.effort`. `thinking.type` accepts enabled/disabled with an
-  optional `budgetTokens` (or native `budget_tokens`). `outputConfig.format` accepts a JSON schema.
-  Model Studio's empty thinking signatures are accepted; supplied signatures are replayed unchanged.
-- **Responses:** `reasoningEffort` → `reasoning.effort`, plus `enableThinking`, `store`,
-  `previousResponseId`, and `conversation`. Omitted `store` retains the API's default (`true`);
-  set it to `false` for client-managed history. `previousResponseId` requires a stored response.
-  Hosted tools are `Alibaba.webSearch()`, `Alibaba.webExtractor()`, and `Alibaba.codeInterpreter()`.
-  Web extraction is used together with web search. Hosted calls/results carry `providerExecuted: true`.
-
-Omitted options preserve provider defaults. Effort values pass through unchanged and accept future
-strings. Qwen 3.8 Chat rejects requests combining a thinking budget with effort.
-
-Package entrypoints are `@opencode/ai/providers/alibaba`, `alibaba/chat`, `alibaba/messages`,
-and `alibaba/responses`. Live recordings cover all three APIs in Singapore; regional URL construction
-is unit-tested for all six regions.
-
 ## Z.AI
 
 `ZAI` uses the standard API. Chat Completions is the default language-model API;

packages/ai/src/cache-policy.ts

@@ -12,7 +12,6 @@
 // count against the four-breakpoint budget; auto only fills remaining slots.
 import { CacheHint, type CachePolicy, type CachePolicyObject } from "./schema/options.js"
 import { LLMRequest, Message, ToolDefinition, type ContentPart, type ToolEntry } from "./schema/messages.js"
-import { effortUpdate } from "./effort-updates.js"
 
 const AUTO: CachePolicyObject = {
   tools: true,
@@ -122,15 +121,9 @@ const markMessages = (
     return markMessageAt(messages, lastIndexOfRole(messages, "user"), hint, budget)
   if (strategy === "latest-assistant")
     return markMessageAt(messages, lastIndexOfRole(messages, "assistant"), hint, budget)
-  let start = messages.length
-  let remaining = strategy.tail
-  while (remaining > 0 && start > 0) {
-    start -= 1
-    if (effortUpdate(messages[start]!) === undefined) remaining -= 1
-  }
+  const start = Math.max(0, messages.length - strategy.tail)
   let next = messages
-  for (let i = start; i < messages.length; i++)
-    if (effortUpdate(messages[i]!) === undefined) next = markMessageAt(next, i, hint, budget)
+  for (let i = start; i < messages.length; i++) next = markMessageAt(next, i, hint, budget)
   return next
 }
 

packages/ai/src/effort-updates.ts

@@ -1,27 +0,0 @@
-// A top-level reasoning effort change invalidates the whole provider prompt cache, so a
-// mid-conversation switch travels as a `Message.effort(...)` marker: protocols with a native
-// per-message update freeze the top-level effort and lower the markers; every other route strips them.
-import { LLMRequest, type EffortPart, type Message } from "./schema/messages.js"
-
-export const effortUpdate = (message: Message): EffortPart | undefined => {
-  if (message.role !== "system" || message.content.length !== 1) return undefined
-  const part = message.content[0]
-  return part.type === "effort" ? part : undefined
-}
-
-export const stripEffortUpdates = (request: LLMRequest) => {
-  const messages = request.messages.filter((message) => effortUpdate(message) === undefined)
-  return messages.length === request.messages.length ? request : LLMRequest.update(request, { messages })
-}
-
-export const applyEffortUpdates = (request: LLMRequest): LLMRequest =>
-  request.model.route.supportsEffortUpdates?.(request) ? request : stripEffortUpdates(request)
-
-// The markers must end at `current`: `revert.ts` never touches `session.model` and forks may select
-// another variant, so on disagreement fall back to a plain top-level change instead of misreporting effort.
-export const resolveEffortUpdates = (request: LLMRequest, current: string | undefined) => {
-  const updates = request.messages.flatMap((message) => effortUpdate(message) ?? [])
-  if (updates.length === 0) return { request, effort: current }
-  if (updates.at(-1)?.effort !== current) return { request: stripEffortUpdates(request), effort: current }
-  return { request, effort: updates[0]?.previous }
-}

packages/ai/src/protocols/alibaba-chat.ts

@@ -1,92 +0,0 @@
-import { Effect, Schema } from "effect"
-import { Protocol } from "../route/protocol.js"
-import type { LanguageModelCompatibility } from "../schema/index.js"
-import { OpenAIChat } from "./openai-chat.js"
-import { JsonObject, ProviderShared } from "./shared.js"
-import { OpenResponsesOptions } from "./utils/open-responses-options.js"
-
-export type ReasoningEffort = OpenResponsesOptions.ReasoningEffort
-
-const Options = Schema.Struct({
-  reasoningEffort: OpenResponsesOptions.Options.fields.reasoningEffort,
-  enableThinking: Schema.optional(Schema.Boolean),
-  thinkingBudget: Schema.optional(Schema.Int),
-  preserveThinking: Schema.optional(Schema.Boolean),
-  clearThinking: Schema.optional(Schema.Boolean),
-  thinking: Schema.optional(
-    Schema.Struct({
-      type: Schema.declare<"adaptive" | "disabled" | (string & {})>(Schema.is(Schema.String)),
-    }),
-  ),
-  toolStream: Schema.optional(Schema.Boolean),
-  parallelToolCalls: OpenResponsesOptions.Options.fields.parallelToolCalls,
-  repetitionPenalty: Schema.optional(Schema.Number),
-  responseFormat: Schema.optional(
-    Schema.Struct({
-      type: Schema.declare<"text" | "json_object" | "json_schema" | (string & {})>(Schema.is(Schema.String)),
-      json_schema: Schema.optional(JsonObject),
-    }),
-  ),
-  enableSearch: Schema.optional(Schema.Boolean),
-  searchOptions: Schema.optional(
-    Schema.Struct({
-      forced_search: Schema.optional(Schema.Boolean),
-      search_strategy: Schema.optional(
-        Schema.declare<"turbo" | "max" | "agent" | "agent_max" | (string & {})>(Schema.is(Schema.String)),
-      ),
-      enable_search_extension: Schema.optional(Schema.Boolean),
-    }),
-  ),
-})
-export type OptionsInput = typeof Options.Type
-
-export const compatibility = {
-  maxTokensField: "max_completion_tokens",
-  supportsStore: false,
-  supportsStrictMode: false,
-  reasoningField: "reasoning_content",
-  zaiToolStream: false,
-} satisfies LanguageModelCompatibility
-
-export const protocol = Protocol.make({
-  id: "alibaba-chat",
-  body: {
-    schema: Schema.Struct({
-      ...OpenAIChat.bodyFields,
-      enable_thinking: Options.fields.enableThinking,
-      thinking_budget: Options.fields.thinkingBudget,
-      preserve_thinking: Options.fields.preserveThinking,
-      clear_thinking: Options.fields.clearThinking,
-      thinking: Options.fields.thinking,
-      parallel_tool_calls: Options.fields.parallelToolCalls,
-      repetition_penalty: Options.fields.repetitionPenalty,
-      top_k: Schema.optional(Schema.Int),
-      response_format: Options.fields.responseFormat,
-      enable_search: Options.fields.enableSearch,
-      search_options: Options.fields.searchOptions,
-    }),
-    from: Effect.fn("AlibabaChat.fromRequest")(function* (req) {
-      const opts = yield* ProviderShared.validateWith(Schema.decodeUnknownEffect(Options))(req.providerOptions ?? {})
-      return {
-        ...(yield* OpenAIChat.protocol.body.from(req)),
-        enable_thinking: opts.enableThinking,
-        thinking_budget: opts.thinkingBudget,
-        preserve_thinking: opts.preserveThinking,
-        clear_thinking: opts.clearThinking,
-        thinking: opts.thinking,
-        tool_stream: opts.toolStream,
-        parallel_tool_calls:
-          opts.parallelToolCalls ??
-          (req.toolChoice?.disableParallelToolUse === undefined ? undefined : !req.toolChoice.disableParallelToolUse),
-        repetition_penalty: opts.repetitionPenalty,
-        top_k: req.generation?.topK,
-        response_format: opts.responseFormat,
-        enable_search: opts.enableSearch,
-        search_options: opts.searchOptions,
-      }
-    }),
-  },
-  stream: OpenAIChat.protocol.stream,
-})
-
-export * as AlibabaChat from "./alibaba-chat.js"

packages/ai/src/protocols/alibaba-messages.ts

@@ -1,48 +0,0 @@
-import { Effect, Schema } from "effect"
-import { Protocol } from "../route/protocol.js"
-import { LLMRequest } from "../schema/index.js"
-import { AnthropicMessages } from "./anthropic-messages.js"
-import { ProviderShared } from "./shared.js"
-import { OpenResponsesOptions } from "./utils/open-responses-options.js"
-
-const Options = Schema.Struct({
-  effort: Schema.optional(OpenResponsesOptions.ReasoningEffort),
-  thinking: Schema.optional(
-    Schema.Struct({
-      type: Schema.declare<"enabled" | "disabled" | (string & {})>(Schema.is(Schema.String)),
-      budgetTokens: Schema.optional(Schema.Int),
-      budget_tokens: Schema.optional(Schema.Int),
-    }),
-  ),
-})
-export type OptionsInput = typeof Options.Type & Pick<AnthropicMessages.OptionsInput, "outputConfig">
-export const protocol = Protocol.make({
-  id: "alibaba-messages",
-  body: {
-    schema: Schema.Struct({
-      ...AnthropicMessages.AnthropicMessagesBody.fields,
-      thinking: Schema.optional(Schema.Struct({ type: Schema.String, budget_tokens: Schema.optional(Schema.Int) })),
-    }),
-    from: Effect.fn("AlibabaMessages.fromRequest")(function* (req) {
-      const opts = yield* ProviderShared.validateWith(Schema.decodeUnknownEffect(Options))(req.providerOptions ?? {})
-      // Model Studio accepts enabled thinking without Anthropic's mandatory token budget.
-      return {
-        ...(yield* AnthropicMessages.protocol.body.from(
-          LLMRequest.update(req, {
-            providerOptions: { ...req.providerOptions, thinking: undefined },
-          }),
-        )),
-        thinking:
-          opts.thinking === undefined
-            ? undefined
-            : {
-                type: opts.thinking.type,
-                budget_tokens: opts.thinking.budgetTokens ?? opts.thinking.budget_tokens,
-              },
-      }
-    }),
-  },
-  stream: AnthropicMessages.protocol.stream,
-})
-
-export * as AlibabaMessages from "./alibaba-messages.js"

packages/ai/src/protocols/alibaba-responses.ts

@@ -1,98 +0,0 @@
-import { Effect, Schema } from "effect"
-import { Protocol } from "../route/protocol.js"
-import { OpenResponses } from "./open-responses.js"
-import { JsonObject, optionalArray, ProviderShared } from "./shared.js"
-import { OpenResponsesOptions } from "./utils/open-responses-options.js"
-import { ResponsesHostedTools } from "./utils/responses-hosted-tools.js"
-
-const Options = Schema.Struct({
-  reasoningEffort: OpenResponsesOptions.Options.fields.reasoningEffort,
-  enableThinking: Schema.optional(Schema.Boolean),
-  store: OpenResponsesOptions.Options.fields.store,
-  previousResponseId: Schema.optional(Schema.String),
-  conversation: Schema.optional(Schema.String),
-})
-export type OptionsInput = typeof Options.Type
-const NativeTool = Schema.Struct({ type: Schema.Literals(["web_search", "web_extractor", "code_interpreter"]) })
-const WebExtractorItem = Schema.StructWithRest(
-  Schema.Struct({
-    type: Schema.Literal("web_extractor_call"),
-    id: Schema.String,
-    urls: Schema.optional(Schema.Array(Schema.String)),
-    goal: Schema.optional(Schema.String),
-  }),
-  [JsonObject],
-)
-const Body = Schema.Struct({
-  ...OpenResponses.coreFields,
-  input: Schema.Array(Schema.Union([OpenResponses.InputItem, WebExtractorItem])),
-  tools: optionalArray(Schema.Union([OpenResponses.Tool, NativeTool])),
-  enable_thinking: Options.fields.enableThinking,
-  previous_response_id: Options.fields.previousResponseId,
-  conversation: Options.fields.conversation,
-  stream: Schema.Literal(true),
-})
-const adapter = {
-  id: "alibaba-responses",
-  name: "Alibaba Responses",
-  nativeTool: (native) => ProviderShared.validateWith(Schema.decodeUnknownEffect(NativeTool))(native.alibaba),
-  restoreHostedToolItem: (item: unknown) => (Schema.is(WebExtractorItem)(item) ? item : undefined),
-} satisfies OpenResponses.ProviderAdapter
-
-const tools = {
-  web_search_call: { name: "web_search", input: (item) => item.action ?? {} },
-  code_interpreter_call: { name: "code_interpreter", input: (item) => ({ code: item.code }) },
-} satisfies ResponsesHostedTools.Definitions
-
-export const protocol = Protocol.make({
-  id: adapter.id,
-  body: {
-    schema: Body,
-    from: Effect.fn("AlibabaResponses.fromRequest")(function* (req) {
-      const opts = yield* ProviderShared.validateWith(Schema.decodeUnknownEffect(Options))(req.providerOptions ?? {})
-      const body = yield* OpenResponses.fromRequestWithAdapter(req, adapter)
-      const choice = body.tool_choice
-      return yield* ProviderShared.validateWith(Schema.decodeUnknownEffect(Body))({
-        ...body,
-        enable_thinking: opts.enableThinking,
-        previous_response_id: opts.previousResponseId,
-        conversation: opts.conversation,
-        // Model Studio expresses named selection through allowed_tools.
-        tool_choice:
-          typeof choice === "object" && choice.type === "function"
-            ? { type: "allowed_tools" as const, mode: "required" as const, tools: [choice] }
-            : choice,
-      })
-    }),
-  },
-  stream: {
-    event: OpenResponses.protocol.stream.event,
-    initial: (req) => OpenResponses.initial(req, adapter),
-    step: (state, input) =>
-      Effect.gen(function* () {
-        const event = OpenResponses.normalize(state, input)
-        if (event.type !== "response.output_item.done" || !event.item) return yield* OpenResponses.step(state, event)
-        if (event.item.type === "web_extractor_call") {
-          const item = yield* Schema.decodeUnknownEffect(WebExtractorItem)(event.item).pipe(
-            Effect.mapError((cause) =>
-              ProviderShared.eventError(
-                adapter.id,
-                "Alibaba returned an invalid web extraction item",
-                ProviderShared.encodeJson(event),
-                cause,
-              ),
-            ),
-          )
-          return yield* ResponsesHostedTools.onDone(state, item, {
-            web_extractor_call: { name: "web_extractor", input: () => ({ urls: item.urls, goal: item.goal }) },
-          })
-        }
-        if (ResponsesHostedTools.isItem(event.item, tools))
-          return yield* ResponsesHostedTools.onDone(state, event.item, tools)
-        return yield* OpenResponses.step(state, event)
-      }),
-    terminal: OpenResponses.terminal,
-  },
-})
-
-export * as AlibabaResponses from "./alibaba-responses.js"

packages/ai/src/protocols/anthropic-messages.ts

@@ -27,7 +27,6 @@ import {
 } from "../schema/index.js"
 import { JsonObject, optionalArray, optionalNull, ProviderShared } from "./shared.js"
 import { classifyProviderFailure } from "../provider-error.js"
-import { effortUpdate, resolveEffortUpdates } from "../effort-updates.js"
 import * as Cache from "./utils/cache.js"
 import { Lifecycle } from "./utils/lifecycle.js"
 import { ToolSchemaProjection } from "./utils/tool-schema.js"
@@ -37,7 +36,6 @@ const ADAPTER = "anthropic-messages"
 export const DEFAULT_BASE_URL = "https://api.anthropic.com/v1"
 export const PATH = "/messages"
 export const DEFAULT_MAX_TOKENS = 32_000
-const DEFAULT_EFFORT = "high"
 
 const SSE_EVENTS = new Set([
   "message",
@@ -288,11 +286,7 @@ type AnthropicToolResultBlock = Schema.Schema.Type<typeof AnthropicToolResultBlo
 const AnthropicMessage = Schema.Union([
   Schema.Struct({ role: Schema.Literal("user"), content: Schema.Array(AnthropicUserBlock) }),
   Schema.Struct({ role: Schema.Literal("assistant"), content: Schema.Array(AnthropicAssistantBlock) }),
-  Schema.Struct({
-    role: Schema.Literal("system"),
-    content: Schema.Array(AnthropicTextBlock),
-    output_config: Schema.optional(Schema.Struct({ effort: Schema.String })),
-  }),
+  Schema.Struct({ role: Schema.Literal("system"), content: Schema.Array(AnthropicTextBlock) }),
 ]).pipe(Schema.toTaggedUnion("role"))
 type AnthropicMessage = Schema.Schema.Type<typeof AnthropicMessage>
 
@@ -883,12 +877,6 @@ const lowerMessages = Effect.fn("AnthropicMessages.lowerMessages")(function* (
 
   for (const [index, message] of request.messages.entries()) {
     if (message.role === "system") {
-      const update = effortUpdate(message)
-      if (update) {
-        // Accepted at any position, so the text-update placement rules do not apply.
-        messages.push({ role: "system", content: [], output_config: { effort: update.effort ?? DEFAULT_EFFORT } })
-        continue
-      }
       if (splitsLocalToolResults(request.messages, index))
         return yield* invalid("Anthropic Messages system updates cannot split a local tool call from its tool result")
       if (supportsNativeSystemUpdates(request) && canUseNativeSystemUpdate(request, index)) {
@@ -1046,11 +1034,18 @@ const resolveOptions = Effect.fn("AnthropicMessages.resolveOptions")(function* (
     ProviderShared.isRecord(rawOutputConfig) && ProviderShared.isRecord(rawOutputConfig.format)
       ? (rawOutputConfig.format as { type: "json_schema"; schema: Record<string, unknown> })
       : undefined
+  const output_config =
+    outputConfigEffort === undefined && outputConfigFormat === undefined
+      ? undefined
+      : {
+          ...(outputConfigEffort === undefined ? {} : { effort: outputConfigEffort }),
+          ...(outputConfigFormat === undefined ? {} : { format: outputConfigFormat }),
+        }
   const thinking = yield* resolveThinking(input?.thinking)
   return {
     thinking: applyThinkingBindingDefault(request.model, thinking),
     effort: outputConfigEffort,
-    format: outputConfigFormat,
+    output_config,
     service_tier,
     metadata,
     container,
@@ -1059,30 +1054,15 @@ const resolveOptions = Effect.fn("AnthropicMessages.resolveOptions")(function* (
   }
 })
 
-// Accept gateway namespaces and Vertex suffixes without treating a snapshot date as a minor version.
-const claudeVersion = (id: string) => {
-  const match = /(?:^|[./])claude-(?<family>[a-z]+)-(?<major>\d+)(?:[.-](?<minor>\d{1,2}))?(?:$|[-:@])/.exec(
-    id.toLowerCase(),
-  )?.groups
-  if (!match) return undefined
-  return { family: match.family, major: Number(match.major), minor: Number(match.minor ?? 0) }
-}
-
 const supportsThinkingBlockBinding = (model: LLMRequest["model"]) => {
   const override = model.compatibility?.supportsThinkingBlockBinding
   if (override !== undefined) return override
-  const version = claudeVersion(model.id)
-  return version !== undefined && (version.major > 5 || (version.major === 5 && version.minor >= 1))
-}
-
-const supportsEffortUpdates = (model: LLMRequest["model"]) => {
-  const override = model.compatibility?.supportsEffortUpdates
-  if (override !== undefined) return override
-  const version = claudeVersion(model.id)
-  if (version === undefined) return false
-  if (version.family === "opus") return version.major >= 5
-  if (version.family !== "fable" && version.family !== "mythos") return false
-  return version.major > 5 || (version.major === 5 && version.minor >= 1)
+  // Accept gateway namespaces and Vertex suffixes without treating a snapshot date as a minor version.
+  const version = /(?:^|[./])claude-[a-z]+-(?<major>\d+)(?:[.-](?<minor>\d{1,2}))?(?:$|[-:@])/i.exec(model.id)?.groups
+  if (!version) return false
+  const major = Number(version.major)
+  const minor = Number(version.minor ?? 0)
+  return major > 5 || (major === 5 && minor >= 1)
 }
 
 const applyThinkingBindingDefault = (model: LLMRequest["model"], thinking: AnthropicThinking | undefined) => {
@@ -1124,15 +1104,13 @@ const fromRequest = Effect.fn("AnthropicMessages.fromRequest")(function* (reques
   const management = yield* ProviderShared.validateWith(
     Schema.decodeUnknownEffect(Schema.UndefinedOr(ContextManagement)),
   )(request.providerOptions?.contextManagement)
-  const options = yield* resolveOptions(request)
-  const updates = resolveEffortUpdates(request, options.effort)
   const generation = request.generation
   const toolSchemaCompatibility = request.model.compatibility?.toolSchema
   // Allocate the 4-breakpoint budget in invalidation order: tools → system →
   // messages. Tools live highest in the cache hierarchy, so when callers
   // over-mark we keep their tool hints and shed the message-tail ones first.
   const breakpoints = Cache.newBreakpoints(ANTHROPIC_BREAKPOINT_CAP)
-  const flattened = ProviderShared.flattenToolRequest(updates.request)
+  const flattened = ProviderShared.flattenToolRequest(request)
   const tools =
     flattened.tools.length === 0
       ? undefined
@@ -1160,13 +1138,7 @@ const fromRequest = Effect.fn("AnthropicMessages.fromRequest")(function* (reques
       `Anthropic Messages: dropped ${breakpoints.dropped} cache breakpoint(s); the API allows at most ${ANTHROPIC_BREAKPOINT_CAP} per request.`,
     )
   }
-  const output_config =
-    updates.effort === undefined && options.format === undefined
-      ? undefined
-      : {
-          ...(updates.effort === undefined ? {} : { effort: updates.effort }),
-          ...(options.format === undefined ? {} : { format: options.format }),
-        }
+  const options = yield* resolveOptions(request)
   const body = {
     model: request.model.id,
     system,
@@ -1180,7 +1152,7 @@ const fromRequest = Effect.fn("AnthropicMessages.fromRequest")(function* (reques
     top_k: generation?.topK,
     stop_sequences: generation?.stop,
     thinking: options.thinking,
-    output_config,
+    output_config: options.output_config,
     // top-level passthrough per SDK MessageCreateParamsBase:4638,4643,4649,4654,4670
     cache_control: options.cache_control,
     container: options.container,
@@ -1705,7 +1677,6 @@ export const protocol = Protocol.make({
     }),
     step,
   },
-  supportsEffortUpdates: (request) => supportsEffortUpdates(request.model),
 })
 
 export const transport = <
@@ -1744,9 +1715,6 @@ function requiredBetaHeaders(body: Pick<AnthropicMessagesBody, "messages" | "con
   )
   if (requestsCompaction || replaysCompaction) betas.push("compact-2026-01-12")
 
-  if (body.messages.some((message) => message.role === "system" && message.output_config !== undefined))
-    betas.push("mid-conversation-output-config-2026-07-01")
-
   const thinking = body.thinking
   if (thinking && thinking.type !== "disabled" && thinking.block_binding)
     betas.push("thinking-binding-controls-2026-08-01")

packages/ai/src/protocols/open-responses-channel.ts

@@ -18,6 +18,7 @@ const WebSocketResponseCreate = Schema.StructWithRest(Schema.Struct({ type: Sche
 ])
 const decodeMessage = ProviderShared.validateWith(Schema.decodeUnknownEffect(WebSocketResponseCreate))
 const encodeMessage = Schema.encodeSync(Schema.fromJsonString(WebSocketResponseCreate))
+const decodeEvent = Schema.decodeUnknownEffect(OpenResponses.protocol.stream.event)
 
 export interface Options {
   readonly id: string
@@ -26,7 +27,6 @@ export interface Options {
   readonly enabled?: (url: string) => boolean
   readonly url?: (url: string) => string
   readonly headers?: (headers: Headers.Headers) => Headers.Headers
-  readonly continuation?: OpenResponsesContinuation.Shape
 }
 
 export interface Prepared {
@@ -60,7 +60,7 @@ const driver = (options: Options, body: string): WebSocketChannelDriver => {
       }),
     observe: (_create, frame) =>
       Effect.gen(function* () {
-        const event = yield* OpenResponses.decodeChannelEvent(frame).pipe(
+        const event = yield* decodeEvent(frame).pipe(
           Effect.mapError((cause) =>
             ProviderShared.eventError(options.id, `Invalid ${options.name} WebSocket event`, frame, cause),
           ),
@@ -163,7 +163,6 @@ export const transport = <Body>(options: Options): Transport<Body, Prepared, str
                     request: create.request,
                     message: create.message,
                     base,
-                    continuation: options.continuation,
                   }),
                 }
               })

packages/ai/src/protocols/open-responses-continuation.ts

@@ -6,6 +6,7 @@ import { OpenResponses } from "./open-responses.js"
 
 const PROTOCOL = "open-responses.websocket.v1"
 const VERSION = 1
+const decodeEvent = Schema.decodeUnknownEffect(OpenResponses.protocol.stream.event)
 
 interface CheckpointValue {
   readonly version: typeof VERSION
@@ -14,19 +15,12 @@ interface CheckpointValue {
   readonly output: ReadonlyArray<unknown>
 }
 
-/**
- * Fields to send next to `previous_response_id` on an incremental step, or undefined to send the step in full.
- * Whether omitted fields carry over from the continued response is provider behavior the route must know.
- */
-export type Shape = (request: Readonly<Record<string, unknown>>) => Readonly<Record<string, unknown>> | undefined
-
 export interface DriverInput {
   readonly id: string
   readonly name: string
   readonly request: Readonly<Record<string, unknown>>
   readonly message: string
   readonly base: WebSocketChannelDriver
-  readonly continuation?: Shape
 }
 
 const checkpointValue = (checkpoint: ChannelCheckpoint | undefined): CheckpointValue | undefined => {
@@ -133,26 +127,22 @@ const rejected = (
 
 export const driver = (input: DriverInput): WebSocketChannelDriver => {
   const { previous_response_id: _previousResponseID, ...request } = input.request
-  const shape = input.continuation ?? ((fields: Readonly<Record<string, unknown>>) => fields)
   let output: OpenResponses.StreamItem[] = []
   return {
     create: (checkpoint) =>
       Effect.sync(() => {
         output = []
         const previous = checkpointValue(checkpoint)
-        // Ask the route first: diffing the whole history is wasted when it declines the continuation.
-        const fields = previous ? shape(request) : undefined
-        const delta = previous && fields ? incremental(request, previous) : undefined
-        if (!previous || !fields || !delta)
-          return { message: ProviderShared.encodeJson(request), mode: "full" as const }
+        const delta = previous ? incremental(request, previous) : undefined
+        if (!previous || !delta) return { message: ProviderShared.encodeJson(request), mode: "full" as const }
         return {
-          message: ProviderShared.encodeJson({ ...fields, input: delta, previous_response_id: previous.responseID }),
+          message: ProviderShared.encodeJson({ ...request, input: delta, previous_response_id: previous.responseID }),
           mode: "incremental" as const,
         }
       }),
     observe: (create, frame) =>
       Effect.gen(function* () {
-        const event = yield* OpenResponses.decodeChannelEvent(frame).pipe(
+        const event = yield* decodeEvent(frame).pipe(
           Effect.mapError((cause) =>
             ProviderShared.eventError(input.id, `Invalid ${input.name} WebSocket event`, frame, cause),
           ),
@@ -163,15 +153,6 @@ export const driver = (input: DriverInput): WebSocketChannelDriver => {
           const rejection = code(event)
           if (rejection === "previous_response_not_found") return rejected(observation, "retry-full")
           if (rejection === "websocket_connection_limit_reached") return rejected(observation, "rotate-and-retry-full")
-          // Only the continuation distinguishes an incremental send from a full one, so an unclassified
-          // invalid request there is retried full; Codex reports a stale previous_response_id that way, with
-          // no code. Classified failures such as context overflow keep their runner-owned recovery.
-          if (
-            create.mode === "incremental" &&
-            observation.error.reason._tag === "InvalidRequest" &&
-            observation.error.reason.classification === undefined
-          )
-            return rejected(observation, "retry-full")
         }
         if (observation.type !== "completed") return observation
         // A trigger installs a different context window. Clear the append baseline, retaining the socket.
@@ -191,7 +172,7 @@ export const driver = (input: DriverInput): WebSocketChannelDriver => {
               responseID,
               request,
               // Completion can re-encrypt reasoning. Callers replay the item already emitted by output_item.done.
-              output: event.response?.output?.length
+              output: event.response?.output
                 ? event.response.output.map((item) =>
                     item.type === "reasoning" && item.id !== undefined
                       ? (output.find((done) => done.type === item.type && done.id === item.id) ?? item)
@@ -205,4 +186,4 @@ export const driver = (input: DriverInput): WebSocketChannelDriver => {
   }
 }
 
-export * as OpenResponsesContinuation from "./open-responses-continuation.js"
+export const OpenResponsesContinuation = { driver } as const

packages/ai/src/protocols/open-responses.ts

@@ -1,4 +1,4 @@
-import { Effect, Option, Schema, SchemaGetter } from "effect"
+import { Effect, Option, Schema } from "effect"
 import type { Content } from "@opencode/schema/tool"
 import { HttpTransport } from "../route/transport/index.js"
 import { Protocol } from "../route/protocol.js"
@@ -20,7 +20,6 @@ import {
 } from "../schema/index.js"
 import { JsonObject, optionalArray, optionalNull, ProviderShared } from "./shared.js"
 import { classifyProviderFailure } from "../provider-error.js"
-import { effortUpdate } from "../effort-updates.js"
 import { OpenResponsesOptions } from "./utils/open-responses-options.js"
 import { Lifecycle } from "./utils/lifecycle.js"
 import { ToolSchemaProjection } from "./utils/tool-schema.js"
@@ -165,13 +164,6 @@ export const CompactionItem = Schema.Struct({
   encrypted_content: Schema.String,
 })
 
-// Kept out of the baseline `InputItem` union: only the OpenAI extension accepts it.
-export const ConfigurationUpdate = Schema.Struct({
-  type: Schema.Literal("configuration_update"),
-  reasoning: Schema.Struct({ effort: OpenResponsesOptions.ReasoningEffort }),
-})
-type ConfigurationUpdate = Schema.Schema.Type<typeof ConfigurationUpdate>
-
 export const InputItem = Schema.Union([
   CompactionItem,
   Schema.Struct({ role: Schema.tag("system"), content: Schema.String }),
@@ -216,7 +208,6 @@ export type HostedToolReplayItem = {
 type LoweredInputItem =
   | OpenResponsesInputItem
   | HostedToolReplayItem
-  | ConfigurationUpdate
   | {
       readonly type: "message"
       readonly id?: string
@@ -334,8 +325,9 @@ export const StreamItem = Schema.StructWithRest(
 export type StreamItem = Schema.Schema.Type<typeof StreamItem>
 export type OutputItem = StreamItem & { readonly id: string }
 
-// Responses-compatible providers put streaming error details at the top level or
-// under `error`, and response failures under `response.error`. Accept all three shapes.
+// The Responses schema puts streaming error details at the top level and
+// response failures under `response.error`. WebSocket failures use an
+// event-level `error` envelope, so accept all three shapes here.
 // https://www.openresponses.org/specification
 const OpenResponsesErrorPayload = Schema.Struct({
   type: optionalNull(Schema.String),
@@ -409,45 +401,13 @@ export const Event = Schema.StructWithRest(
     headers: Schema.optional(Schema.Unknown),
   }),
   [Schema.Record(Schema.String, Schema.Unknown)],
-).pipe(
-  Schema.decode({
-    decode: SchemaGetter.transform((event) => {
-      if (event.type !== "error" || event.error != null) return event
-      const { code, message, param, ...rest } = event
-      if (code === undefined && message === undefined && param === undefined) return event
-      // Flat errors (for example, Meta's) can also arrive through generic Responses endpoints.
-      return { ...rest, error: { code, message, param } }
-    }),
-    encode: SchemaGetter.passthrough(),
-  }),
 )
 export type Event = Schema.Schema.Type<typeof Event>
 export type NormalizedEvent = Event & { readonly item?: OutputItem | null }
 
-const decodeEventValue = Schema.decodeUnknownEffect(Event)
-const decodeFrame = Schema.decodeUnknownEffect(ProviderShared.Json)
-
-/**
- * Decodes one WebSocket frame. xAI answers a rejected `response.create` with `{ "error": { "message", "type" } }` and no
- * event type; that envelope reads as an error event so the failure classifies instead of failing decoding.
- */
-export const decodeChannelEvent = (frame: string) =>
-  decodeFrame(frame).pipe(
-    Effect.flatMap((value) =>
-      decodeEventValue(
-        ProviderShared.isRecord(value) && value.type === undefined && ProviderShared.isRecord(value.error)
-          ? { ...value, type: "error" }
-          : value,
-      ),
-    ),
-  )
-
 export interface ProviderAdapter {
   readonly id: string
   readonly name: string
-  readonly nativeTool?: (
-    native: NonNullable<ToolDefinition["native"]>,
-  ) => Effect.Effect<{ readonly type: string }, AIError>
   readonly lowerMedia?: (input: {
     readonly part: MediaPart
     readonly media: ProviderShared.NormalizedMedia
@@ -643,8 +603,6 @@ const lowerToolResultOutput = Effect.fnUntraced(function* (
   return yield* Effect.forEach(content, (item) => lowerToolResultContentItem(item, request, adapter))
 })
 
-const DEFAULT_EFFORT = "medium"
-
 const lowerMessages = Effect.fn("OpenResponses.lowerMessages")(function* (
   request: LLMRequest,
   adapter: ProviderAdapter,
@@ -657,14 +615,6 @@ const lowerMessages = Effect.fn("OpenResponses.lowerMessages")(function* (
       Schema.decodeUnknownEffect(Schema.UndefinedOr(MessageMetadata)),
     )(message.providerMetadata?.[providerMetadataKey])
     if (message.role === "system") {
-      const update = effortUpdate(message)
-      if (update) {
-        // Consecutive updates are rejected, so a newer one replaces its predecessor.
-        const last = input.at(-1)
-        if (last !== undefined && "type" in last && last.type === "configuration_update") input.pop()
-        input.push({ type: "configuration_update", reasoning: { effort: update.effort ?? DEFAULT_EFFORT } })
-        continue
-      }
       input.push({
         role: "developer",
         content: ProviderShared.joinText(yield* ProviderShared.systemUpdateText(adapter.name, message)),
@@ -702,8 +652,7 @@ const lowerMessages = Effect.fn("OpenResponses.lowerMessages")(function* (
             type: "message" as const,
             ...(group.id === undefined ? {} : { id: group.id }),
             role: "assistant" as const,
-            // Replayed text is a finished input item, even if generation was cut short.
-            status: "completed",
+            status: metadata?.status,
             content: group.parts.map((part) => ({ type: "output_text" as const, text: part.text })),
             ...(group.phase === undefined ? {} : { phase: group.phase }),
           })),
@@ -808,7 +757,8 @@ export const lowerConversation = Effect.fn("OpenResponses.lowerConversation")(fu
   }
 })
 
-export const lowerGeneration = (request: LLMRequest, options = OpenResponsesOptions.resolve(request)) => {
+export const lowerGeneration = (request: LLMRequest) => {
+  const options = OpenResponsesOptions.resolve(request)
   const generation = request.generation
   const cacheKey = ProviderShared.promptCacheKey(request)
   const parallelToolCalls = resolveParallelToolCalls(request)
@@ -869,13 +819,11 @@ export const fromRequestWithAdapter = Effect.fn("OpenResponses.fromRequestWithAd
       projected.tools.length === 0
         ? undefined
         : yield* Effect.forEach(projected.tools, (tool) =>
-            tool.native !== undefined && adapter.nativeTool
-              ? adapter.nativeTool(tool.native)
-              : lowerTool(
-                  adapter.name,
-                  tool,
-                  ToolSchemaProjection.modelCompatibility(tool.inputSchema, toolSchemaCompatibility),
-                ),
+            lowerTool(
+              adapter.name,
+              tool,
+              ToolSchemaProjection.modelCompatibility(tool.inputSchema, toolSchemaCompatibility),
+            ),
           ),
     tool_choice:
       allowedToolChoice(request) ??

packages/ai/src/protocols/openai-responses.ts

@@ -6,9 +6,7 @@ import { Endpoint } from "../route/endpoint.js"
 import { Protocol } from "../route/protocol.js"
 import { HttpTransport } from "../route/transport/index.js"
 import { LLMRequest, mergeJsonRecords, type JsonSchema, type ToolDefinition, type ToolEntry } from "../schema/index.js"
-import { resolveEffortUpdates } from "../effort-updates.js"
 import { OpenResponses } from "./open-responses.js"
-import { OpenResponsesOptions } from "./utils/open-responses-options.js"
 import { JsonObject, optionalArray, optionalNull, ProviderShared } from "./shared.js"
 import { OpenAIImage } from "./utils/openai-image.js"
 import { ResponsesHostedTools } from "./utils/responses-hosted-tools.js"
@@ -96,15 +94,9 @@ const OpenAIResponsesToolChoice = Schema.Union([
   Schema.Struct({ type: Schema.tag("image_generation") }),
 ])
 
-const OpenAIResponsesInputItem = Schema.Union([
-  OpenResponses.InputItem,
-  OpenAIResponsesHostedToolItem,
-  OpenResponses.ConfigurationUpdate,
-])
-
 const OpenAIResponsesCoreFields = {
   ...OpenResponses.coreFields,
-  input: Schema.Array(OpenAIResponsesInputItem),
+  input: Schema.Array(Schema.Union([OpenResponses.InputItem, OpenAIResponsesHostedToolItem])),
   tools: optionalArray(OpenAIResponsesTools),
   tool_choice: Schema.optional(OpenAIResponsesToolChoice),
   context_management: Schema.optional(
@@ -127,7 +119,7 @@ export type OpenAIResponsesBody = Schema.Schema.Type<typeof OpenAIResponsesBody>
 export const CompactionTrigger = Schema.Struct({ type: Schema.Literal("compaction_trigger") })
 const CheckpointBody = Schema.Struct({
   ...OpenAIResponsesBody.fields,
-  input: Schema.Array(Schema.Union([OpenAIResponsesInputItem, CompactionTrigger])),
+  input: Schema.Array(Schema.Union([OpenResponses.InputItem, OpenAIResponsesHostedToolItem, CompactionTrigger])),
   store: Schema.Literal(false),
   prompt_cache_retention: optionalNull(Schema.String),
   prompt_cache_options: optionalNull(
@@ -141,14 +133,6 @@ const adapter = {
   restoreHostedToolItem: (item: unknown) => (Schema.is(OpenAIResponsesHostedToolItem)(item) ? item : undefined),
 } satisfies OpenResponses.ProviderAdapter
 
-// Only GPT-6 Astra accepts `configuration_update`, and never alongside automatic `context_management` compaction.
-const supportsEffortUpdates = (request: LLMRequest) => {
-  if (request.providerOptions?.contextManagement !== undefined) return false
-  const override = request.model.compatibility?.supportsEffortUpdates
-  if (override !== undefined) return override
-  return /(?:^|\/)gpt-6-astra$/i.test(request.model.id)
-}
-
 const nativeImageToolInput = (tool: ToolDefinition) => {
   const native = tool.native?.openai
   return ProviderShared.isRecord(native) && native.type === "image_generation" ? native : undefined
@@ -205,12 +189,10 @@ const fromRequest = Effect.fn("OpenAIResponses.fromRequest")(function* (request:
   const management = yield* ProviderShared.validateWith(
     Schema.decodeUnknownEffect(Schema.UndefinedOr(ContextManagement)),
   )(request.providerOptions?.contextManagement)
-  const options = OpenResponsesOptions.resolve(request)
-  const updates = resolveEffortUpdates(request, options.reasoningEffort)
   const toolSchemaCompatibility = request.model.compatibility?.toolSchema
   return yield* decodeBody({
-    ...(yield* OpenResponses.lowerConversation(updates.request, adapter)),
-    ...OpenResponses.lowerGeneration(request, { ...options, reasoningEffort: updates.effort }),
+    ...(yield* OpenResponses.lowerConversation(request, adapter)),
+    ...OpenResponses.lowerGeneration(request),
     context_management: management?.map((edit) => ({ type: edit.type, compact_threshold: edit.compactThreshold })),
     tools:
       request.tools.length === 0
@@ -313,7 +295,6 @@ export const protocol = Protocol.make({
     step,
     terminal: OpenResponses.terminal,
   },
-  supportsEffortUpdates,
 })
 
 const endpoint = Endpoint.path<OpenAIResponsesBody>(PATH, { baseURL: DEFAULT_BASE_URL })

packages/ai/src/protocols/utils/open-responses-options.ts

@@ -1,7 +1,12 @@
 import { Option, Schema } from "effect"
-import { ReasoningEffort, ReasoningEfforts, type LLMRequest } from "../../schema/index.js"
+import type { LLMRequest } from "../../schema/index.js"
 
-export { ReasoningEffort, ReasoningEfforts }
+export const ReasoningEfforts = ["none", "minimal", "low", "medium", "high", "xhigh", "max"] as const
+export type ReasoningEffort = (typeof ReasoningEfforts)[number] | (string & {})
+export const ReasoningEffort = Schema.declare<ReasoningEffort>(
+  (value): value is ReasoningEffort => typeof value === "string",
+  { title: "ReasoningEffort" },
+)
 
 export const TextVerbosities = ["low", "medium", "high"] as const
 export type TextVerbosity = (typeof TextVerbosities)[number] | (string & {})

packages/ai/src/protocols/utils/responses-compaction.ts

@@ -11,7 +11,6 @@ import {
   mergeJsonRecords,
 } from "../../schema/index.js"
 import type { CompactOperation } from "../../route/client.js"
-import { stripEffortUpdates } from "../../effort-updates.js"
 import { Endpoint } from "../../route/endpoint.js"
 import { RequestExecutor } from "../../route/executor.js"
 import { HttpTransport } from "../../route/transport/index.js"
@@ -76,8 +75,7 @@ const Response = Schema.Struct({
 export const make = (adapter: OpenResponses.ProviderAdapter): CompactOperation =>
   Effect.fn("ResponsesCompaction.execute")(function* (request, executor, options) {
     const route = request.model.route
-    // The standalone compaction endpoint rejects histories containing configuration updates.
-    const native = yield* OpenResponses.lowerConversation(stripEffortUpdates(request), adapter)
+    const native = yield* OpenResponses.lowerConversation(request, adapter)
     const body = yield* ProviderShared.validateWith(Schema.decodeUnknownEffect(Body))(
       mergeJsonRecords(
         {

packages/ai/src/providers/alibaba.ts

@@ -1,139 +0,0 @@
-import type { ProviderPackage } from "../provider-package.js"
-import { AlibabaChat } from "../protocols/alibaba-chat.js"
-import { AlibabaMessages } from "../protocols/alibaba-messages.js"
-import { AlibabaResponses } from "../protocols/alibaba-responses.js"
-import { AuthOptions, type AtLeastOne, type ProviderAuthOption } from "../route/auth-options.js"
-import { Route, type RouteDefaultsInput } from "../route/client.js"
-import { Endpoint } from "../route/endpoint.js"
-import { Framing } from "../route/framing.js"
-import { ProviderConfigurationError, ProviderID, ToolDefinition, type ModelID } from "../schema/index.js"
-
-export const id = ProviderID.make("alibaba")
-
-export type Region =
-  | "ap-southeast-1"
-  | "cn-beijing"
-  | "cn-hongkong"
-  | "us-east-1"
-  | "eu-central-1"
-  | "ap-northeast-1"
-  | (string & {})
-export type ChatOptionsInput = AlibabaChat.OptionsInput
-export type MessagesOptionsInput = AlibabaMessages.OptionsInput
-export type ResponsesOptionsInput = AlibabaResponses.OptionsInput
-
-type Location = AtLeastOne<{
-  readonly region: Region
-  /** Overrides the selected API's complete base URL, including its version prefix. */
-  readonly baseURL: string
-}> & { readonly workspaceID?: string }
-
-export type Config = Location &
-  Omit<RouteDefaultsInput, "providerOptions"> &
-  ProviderAuthOption<"optional"> & {
-    readonly providerOptions?: ChatOptionsInput | MessagesOptionsInput | ResponsesOptionsInput
-  }
-export type Settings<Options = ChatOptionsInput> = Location &
-  ProviderPackage.Settings & {
-    readonly apiKey?: string
-    readonly providerOptions?: Options
-  }
-
-const hosts = new Map<string, string>([
-  ["ap-southeast-1", "dashscope-intl.aliyuncs.com"],
-  ["cn-beijing", "dashscope.aliyuncs.com"],
-  ["cn-hongkong", "cn-hongkong.dashscope.aliyuncs.com"],
-  ["us-east-1", "dashscope-us.aliyuncs.com"],
-])
-const chatRoute = Route.make({
-  id: "alibaba-chat",
-  provider: id,
-  providerMetadataKey: "alibaba",
-  protocol: AlibabaChat.protocol,
-  endpoint: Endpoint.path("/chat/completions"),
-  framing: Framing.sse,
-})
-const messagesRoute = Route.make({
-  id: "alibaba-messages",
-  provider: id,
-  providerMetadataKey: "alibaba",
-  protocol: AlibabaMessages.protocol,
-  endpoint: Endpoint.path("/messages"),
-  framing: Framing.sse,
-  headers: () => ({ "anthropic-version": "2023-06-01" }),
-})
-const responsesRoute = Route.make({
-  id: "alibaba-responses",
-  provider: id,
-  providerMetadataKey: "alibaba",
-  protocol: AlibabaResponses.protocol,
-  endpoint: Endpoint.path("/responses"),
-  framing: Framing.sse,
-})
-
-export const routes = [chatRoute, messagesRoute, responsesRoute]
-
-export const configure = (input: Config) => {
-  const { apiKey: _key, auth: _auth, region, workspaceID, baseURL, ...rest } = input
-  const host =
-    region === undefined
-      ? undefined
-      : workspaceID === undefined
-        ? hosts.get(region)
-        : `${workspaceID}.${region}.maas.aliyuncs.com`
-  if (baseURL === undefined) {
-    if (region === undefined)
-      throw new ProviderConfigurationError({ provider: id, message: "Alibaba requires region or baseURL" })
-    if (host === undefined)
-      throw new ProviderConfigurationError({
-        provider: id,
-        message: `Alibaba region ${region} requires workspaceID or baseURL`,
-      })
-  }
-  const opts = { ...rest, auth: AuthOptions.bearer(input, ["DASHSCOPE_API_KEY", "ALIBABA_API_KEY"]) }
-  const common = { ...opts, endpoint: { baseURL: baseURL ?? `https://${host}/compatible-mode/v1` } }
-  const chat = (id: string | ModelID) =>
-    chatRoute.with(common).model<ChatOptionsInput>({ id, compatibility: AlibabaChat.compatibility })
-  const messages = (id: string | ModelID) =>
-    messagesRoute
-      .with({
-        ...opts,
-        endpoint: { baseURL: baseURL ?? `https://${host}/apps/anthropic/v1` },
-      })
-      .model<MessagesOptionsInput>({ id, compatibility: { requireSignature: false } })
-  const responses = (id: string | ModelID) => responsesRoute.with(common).model<ResponsesOptionsInput>({ id })
-  return { id, model: chat, chat, messages, responses, configure }
-}
-
-export const provider = { id, configure }
-
-export const model: ProviderPackage.Definition<Settings, ChatOptionsInput>["model"] = (id, input) =>
-  fromSettings(input).chat(id)
-export const messagesModel: ProviderPackage.Definition<
-  Settings<MessagesOptionsInput>,
-  MessagesOptionsInput
->["model"] = (id, input) => fromSettings(input).messages(id)
-export const responsesModel: ProviderPackage.Definition<
-  Settings<ResponsesOptionsInput>,
-  ResponsesOptionsInput
->["model"] = (id, input) => fromSettings(input).responses(id)
-
-function fromSettings(input: Settings<Config["providerOptions"]>) {
-  const { body, ...rest } = input
-  return configure({ ...rest, http: body === undefined ? undefined : { body } })
-}
-
-export const webSearch = () => hostedTool("web_search", "Search the web with Alibaba's hosted search tool.")
-export const webExtractor = () => hostedTool("web_extractor", "Extract web page content with Alibaba's hosted tool.")
-export const codeInterpreter = () => hostedTool("code_interpreter", "Execute code with Alibaba's hosted interpreter.")
-
-function hostedTool(type: "web_search" | "web_extractor" | "code_interpreter", description: string) {
-  return ToolDefinition.make({
-    name: type,
-    description,
-    inputSchema: { type: "object", properties: {} },
-    native: { alibaba: { type } },
-  })
-}
-
-export * as Alibaba from "./alibaba.js"

packages/ai/src/providers/alibaba/chat.ts

@@ -1 +0,0 @@
-export { model, type Settings } from "../alibaba.js"

packages/ai/src/providers/alibaba/messages.ts

@@ -1,3 +0,0 @@
-import type { Alibaba } from "../alibaba.js"
-export { messagesModel as model } from "../alibaba.js"
-export type Settings = Alibaba.Settings<Alibaba.MessagesOptionsInput>

packages/ai/src/providers/alibaba/responses.ts

@@ -1,3 +0,0 @@
-import type { Alibaba } from "../alibaba.js"
-export { responsesModel as model } from "../alibaba.js"
-export type Settings = Alibaba.Settings<Alibaba.ResponsesOptionsInput>

packages/ai/src/providers/amazon-bedrock-mantle.ts

@@ -1,10 +1,9 @@
 import { Route, type RouteDefaultsInput } from "../route/client.js"
-import { Endpoint } from "../route/endpoint.js"
 import type { ProviderPackage } from "../provider-package.js"
 import { OpenAIChat } from "../protocols/openai-chat.js"
-import { OpenResponses } from "../protocols/open-responses.js"
+import { OpenAIResponses } from "../protocols/openai-responses.js"
 import { BedrockAuth, type Credentials } from "../protocols/utils/bedrock-auth.js"
-import { ProviderConfigurationError, ProviderID, type ModelID } from "../schema/index.js"
+import { ProviderID, type ModelID } from "../schema/index.js"
 import { withOpenAIOptions, type OpenAIProviderOptionsInput } from "./openai-options.js"
 
 export const id = ProviderID.make("amazon-bedrock")
@@ -38,10 +37,11 @@ const responsesRoute = Route.make({
   id: "bedrock-mantle-responses",
   provider: id,
   providerMetadataKey: "mantle",
-  protocol: OpenResponses.protocol,
-  endpoint: Endpoint.path(OpenResponses.PATH),
-  transport: OpenResponses.httpTransport,
-  defaults: { providerOptions: { store: false, include: ["reasoning.encrypted_content"] } },
+  protocol: OpenAIResponses.protocol,
+  endpoint: OpenAIResponses.route.endpoint,
+  auth: OpenAIResponses.route.auth,
+  transport: OpenAIResponses.httpTransport,
+  defaults: OpenAIResponses.route.defaults,
 })
 
 const chatRoute = OpenAIChat.route.with({
@@ -79,12 +79,9 @@ const defaults = (input: Config) => {
 
 export const configure = (input: Config = {}) => {
   if (input.auth === "bearer" && input.apiKey === undefined && process.env.AWS_BEARER_TOKEN_BEDROCK === undefined)
-    throw new ProviderConfigurationError({ provider: id, message: "Amazon Bedrock Mantle bearer auth requires apiKey" })
+    throw new Error("Amazon Bedrock Mantle bearer auth requires apiKey")
   if (input.auth === "sigv4" && input.apiKey !== undefined)
-    throw new ProviderConfigurationError({
-      provider: id,
-      message: "Amazon Bedrock Mantle SigV4 auth does not accept apiKey",
-    })
+    throw new Error("Amazon Bedrock Mantle SigV4 auth does not accept apiKey")
   const configuredResponsesRoute = configuredRoute(responsesRoute, input)
   const configuredChatRoute = configuredRoute(chatRoute, input)
   const modelDefaults = defaults(input)

packages/ai/src/providers/amazon-bedrock.ts

@@ -1,6 +1,6 @@
 import type { RouteDefaultsInput } from "../route/client.js"
 import type { ProviderPackage } from "../provider-package.js"
-import { ProviderConfigurationError, ProviderID, type ModelID } from "../schema/index.js"
+import { ProviderID, type ModelID } from "../schema/index.js"
 import * as BedrockConverse from "../protocols/bedrock-converse.js"
 import type { BedrockCredentials } from "../protocols/bedrock-converse.js"
 import { BedrockAuth } from "../protocols/utils/bedrock-auth.js"
@@ -39,9 +39,8 @@ const bedrockBaseURL = (region: string) => `https://bedrock-runtime.${region}.am
 const configuredRoute = (input: Config) => {
   const { apiKey, auth, credentials, profile, region, baseURL, ...rest } = input
   if (auth === "bearer" && apiKey === undefined && process.env.AWS_BEARER_TOKEN_BEDROCK === undefined)
-    throw new ProviderConfigurationError({ provider: id, message: "Amazon Bedrock bearer auth requires apiKey" })
-  if (auth === "sigv4" && apiKey !== undefined)
-    throw new ProviderConfigurationError({ provider: id, message: "Amazon Bedrock SigV4 auth does not accept apiKey" })
+    throw new Error("Amazon Bedrock bearer auth requires apiKey")
+  if (auth === "sigv4" && apiKey !== undefined) throw new Error("Amazon Bedrock SigV4 auth does not accept apiKey")
   const resolvedRegion = BedrockAuth.resolveRegion(input)
   return BedrockConverse.route.with({
     ...rest,

packages/ai/src/providers/anthropic-compatible.ts

@@ -3,7 +3,7 @@ import { AnthropicMessages } from "../protocols/anthropic-messages.js"
 import { Auth } from "../route/auth.js"
 import type { ProviderAuthOption } from "../route/auth-options.js"
 import type { RouteDefaultsInput } from "../route/client.js"
-import { ProviderConfigurationError, ProviderID, type ModelID } from "../schema/index.js"
+import { ProviderID, type ModelID } from "../schema/index.js"
 
 export type AnthropicOptionsInput = AnthropicMessages.OptionsInput
 export type AnthropicProviderOptionsInput = AnthropicMessages.ProviderOptionsInput
@@ -36,12 +36,8 @@ const auth = (input: ProviderAuthOption<"optional">) => {
 }
 
 export const configure = (input: Config) => {
+  if (!input.baseURL) throw new Error("Anthropic-compatible providers require a baseURL")
   const provider = input.provider ?? "anthropic-compatible"
-  if (!input.baseURL)
-    throw new ProviderConfigurationError({
-      provider: ProviderID.make(provider),
-      message: "Anthropic-compatible providers require a baseURL",
-    })
   const { provider: _, baseURL, apiKey: _apiKey, auth: _auth, ...rest } = input
   const route = AnthropicMessages.route.with({
     ...rest,
@@ -65,13 +61,8 @@ export const model: ProviderPackage.Definition<Settings, AnthropicMessages.Provi
   modelID,
   settings,
 ) => {
-  // Read before the exclusivity check narrows a conflicting settings object to `never`.
-  const provider = ProviderID.make(settings.provider ?? id)
   if (settings.apiKey !== undefined && settings.authToken !== undefined)
-    throw new ProviderConfigurationError({
-      provider,
-      message: "Anthropic-compatible apiKey cannot be combined with authToken",
-    })
+    throw new Error("Anthropic-compatible apiKey cannot be combined with authToken")
   return configure({
     ...(settings.authToken === undefined ? { apiKey: settings.apiKey } : { auth: Auth.bearer(settings.authToken) }),
     baseURL: settings.baseURL,

packages/ai/src/providers/anthropic.ts

@@ -2,7 +2,7 @@ import type { RouteDefaultsInput } from "../route/client.js"
 import { Auth } from "../route/auth.js"
 import type { ProviderAuthOption } from "../route/auth-options.js"
 import type { ProviderPackage } from "../provider-package.js"
-import { ProviderConfigurationError, ProviderID, type ModelID } from "../schema/index.js"
+import { ProviderID, type ModelID } from "../schema/index.js"
 import { AnthropicMessages } from "../protocols/anthropic-messages.js"
 import { AnthropicCompatible } from "./anthropic-compatible.js"
 
@@ -57,10 +57,7 @@ export const model: ProviderPackage.Definition<Settings, AnthropicMessages.Provi
   settings,
 ) => {
   if (settings.apiKey !== undefined && settings.authToken !== undefined)
-    throw new ProviderConfigurationError({
-      provider: id,
-      message: "Anthropic apiKey cannot be combined with authToken",
-    })
+    throw new Error("Anthropic apiKey cannot be combined with authToken")
   return configure({
     ...(settings.authToken === undefined ? { apiKey: settings.apiKey } : { auth: Auth.bearer(settings.authToken) }),
     baseURL: settings.baseURL,

packages/ai/src/providers/azure.ts

@@ -3,7 +3,7 @@ import { Auth } from "../route/auth.js"
 import { type AtLeastOne, type ProviderAuthOption } from "../route/auth-options.js"
 import type { Route, RouteDefaultsInput, CompactionOperations } from "../route/client.js"
 import type { ProviderPackage } from "../provider-package.js"
-import { ProviderConfigurationError, ProviderID, type ModelID } from "../schema/index.js"
+import { ProviderID, type ModelID } from "../schema/index.js"
 import * as OpenAIChat from "../protocols/openai-chat.js"
 import * as OpenAIResponses from "../protocols/openai-responses.js"
 import { ProviderShared } from "../protocols/shared.js"
@@ -163,7 +163,7 @@ const config = (settings: Settings): Config => {
   }
   if (settings.baseURL !== undefined) return { ...common, baseURL: settings.baseURL }
   if (settings.resourceName !== undefined) return { ...common, resourceName: settings.resourceName }
-  throw new ProviderConfigurationError({ provider: id, message: "Azure requires resourceName or baseURL" })
+  throw new Error("Azure requires resourceName or baseURL")
 }
 
 export const responsesModel: ProviderPackage.Definition<

packages/ai/src/providers/cloudflare-ai-gateway.ts

@@ -5,7 +5,7 @@ import { Auth } from "../route/auth.js"
 import type { AtLeastOne, ProviderAuthOption } from "../route/auth-options.js"
 import { Route, type RouteDefaultsInput } from "../route/client.js"
 import { Endpoint } from "../route/endpoint.js"
-import { ProviderConfigurationError, ProviderID, type ModelID } from "../schema/index.js"
+import { ProviderID, type ModelID } from "../schema/index.js"
 import type { OpenAIProviderOptionsInput } from "./openai-options.js"
 
 export const id = ProviderID.make("cloudflare-ai-gateway")
@@ -35,11 +35,7 @@ export type Settings = ProviderPackage.Settings &
 
 export const baseURL = (input: GatewayURL) => {
   if (input.baseURL) return input.baseURL
-  if (!input.accountId)
-    throw new ProviderConfigurationError({
-      provider: id,
-      message: "CloudflareAIGateway.configure requires accountId unless baseURL is supplied",
-    })
+  if (!input.accountId) throw new Error("CloudflareAIGateway.configure requires accountId unless baseURL is supplied")
   return `https://gateway.ai.cloudflare.com/v1/${encodeURIComponent(input.accountId)}/${encodeURIComponent(input.gatewayId?.trim() || "default")}/compat`
 }
 

packages/ai/src/providers/cloudflare-workers-ai.ts

@@ -3,7 +3,7 @@ import { OpenAIChat } from "../protocols/openai-chat.js"
 import { AuthOptions, type AtLeastOne, type ProviderAuthOption } from "../route/auth-options.js"
 import { Route, type RouteDefaultsInput } from "../route/client.js"
 import { Endpoint } from "../route/endpoint.js"
-import { ProviderConfigurationError, ProviderID, type ModelID } from "../schema/index.js"
+import { ProviderID, type ModelID } from "../schema/index.js"
 import type { OpenAIProviderOptionsInput } from "./openai-options.js"
 
 export const id = ProviderID.make("cloudflare-workers-ai")
@@ -28,11 +28,7 @@ export type Settings = ProviderPackage.Settings &
 
 export const baseURL = (input: WorkersAIURL) => {
   if (input.baseURL) return input.baseURL
-  if (!input.accountId)
-    throw new ProviderConfigurationError({
-      provider: id,
-      message: "CloudflareWorkersAI.configure requires accountId unless baseURL is supplied",
-    })
+  if (!input.accountId) throw new Error("CloudflareWorkersAI.configure requires accountId unless baseURL is supplied")
   return `https://api.cloudflare.com/client/v4/accounts/${encodeURIComponent(input.accountId)}/ai/v1`
 }
 

packages/ai/src/providers/google-vertex-chat.ts

@@ -2,7 +2,7 @@ import type { ProviderPackage } from "../provider-package.js"
 import { OpenAIChat } from "../protocols/openai-chat.js"
 import { Route, type RouteDefaultsInput } from "../route/client.js"
 import { Endpoint } from "../route/endpoint.js"
-import { ProviderConfigurationError, ProviderID, type ModelID } from "../schema/index.js"
+import { ProviderID, type ModelID } from "../schema/index.js"
 import { GoogleVertexShared } from "./google-vertex-shared.js"
 import type { OpenAIProviderOptionsInput } from "./openai-options.js"
 
@@ -37,8 +37,7 @@ const route = Route.make({
 export const routes = [route]
 
 const configuredRoute = (input: Config) => {
-  if ("apiKey" in input && input.apiKey !== undefined)
-    throw new ProviderConfigurationError({ provider: id, message: "Google Vertex Chat does not support API keys" })
+  if ("apiKey" in input && input.apiKey !== undefined) throw new Error("Google Vertex Chat does not support API keys")
   const {
     accessToken: _accessToken,
     auth: _auth,
@@ -75,8 +74,7 @@ export const provider = {
 }
 
 export const model: ProviderPackage.Definition<Settings, OpenAIProviderOptionsInput>["model"] = (modelID, settings) => {
-  if (settings.apiKey !== undefined)
-    throw new ProviderConfigurationError({ provider: id, message: "Google Vertex Chat does not support API keys" })
+  if (settings.apiKey !== undefined) throw new Error("Google Vertex Chat does not support API keys")
   return configure({
     accessToken: settings.accessToken,
     baseURL: settings.baseURL,

packages/ai/src/providers/google-vertex-messages.ts

@@ -5,7 +5,7 @@ import { Auth } from "../route/auth.js"
 import { Route, type RouteDefaultsInput } from "../route/client.js"
 import { Endpoint } from "../route/endpoint.js"
 import { Protocol } from "../route/protocol.js"
-import { ProviderConfigurationError, ProviderID, type ModelID } from "../schema/index.js"
+import { ProviderID, type ModelID } from "../schema/index.js"
 import { GoogleVertexShared } from "./google-vertex-shared.js"
 
 export type AnthropicOptionsInput = AnthropicMessages.OptionsInput
@@ -67,7 +67,7 @@ export const routes = [route]
 
 const configuredRoute = (input: Config) => {
   if ("apiKey" in input && input.apiKey !== undefined)
-    throw new ProviderConfigurationError({ provider: id, message: "Google Vertex Messages does not support API keys" })
+    throw new Error("Google Vertex Messages does not support API keys")
   const {
     accessToken: _accessToken,
     auth: _auth,
@@ -107,8 +107,7 @@ export const model: ProviderPackage.Definition<Settings, AnthropicMessages.Provi
   modelID,
   settings,
 ) => {
-  if (settings.apiKey !== undefined)
-    throw new ProviderConfigurationError({ provider: id, message: "Google Vertex Messages does not support API keys" })
+  if (settings.apiKey !== undefined) throw new Error("Google Vertex Messages does not support API keys")
   return configure({
     accessToken: settings.accessToken,
     baseURL: settings.baseURL,

packages/ai/src/providers/google-vertex-responses.ts

@@ -2,7 +2,7 @@ import type { ProviderPackage } from "../provider-package.js"
 import { OpenResponses } from "../protocols/open-responses.js"
 import { Route, type RouteDefaultsInput } from "../route/client.js"
 import { Endpoint } from "../route/endpoint.js"
-import { ProviderConfigurationError, ProviderID, type ModelID } from "../schema/index.js"
+import { ProviderID, type ModelID } from "../schema/index.js"
 import { GoogleVertexShared } from "./google-vertex-shared.js"
 import type { OpenResponsesProviderOptionsInput } from "./open-responses-options.js"
 
@@ -39,7 +39,7 @@ export const routes = [route]
 
 const configuredRoute = (input: Config) => {
   if ("apiKey" in input && input.apiKey !== undefined)
-    throw new ProviderConfigurationError({ provider: id, message: "Google Vertex Responses does not support API keys" })
+    throw new Error("Google Vertex Responses does not support API keys")
   const {
     accessToken: _accessToken,
     auth: _auth,
@@ -79,8 +79,7 @@ export const model: ProviderPackage.Definition<Settings, OpenResponsesProviderOp
   modelID,
   settings,
 ) => {
-  if (settings.apiKey !== undefined)
-    throw new ProviderConfigurationError({ provider: id, message: "Google Vertex Responses does not support API keys" })
+  if (settings.apiKey !== undefined) throw new Error("Google Vertex Responses does not support API keys")
   return configure({
     accessToken: settings.accessToken,
     baseURL: settings.baseURL,

packages/ai/src/providers/google-vertex-shared.ts

@@ -1,10 +1,8 @@
 import type { AnyAuthClient } from "google-auth-library"
 import { Effect, Redacted } from "effect"
 import { Auth, MissingCredentialError } from "../route/auth.js"
-import { ProviderConfigurationError, ProviderID } from "../schema/index.js"
 
 const SCOPE = "https://www.googleapis.com/auth/cloud-platform"
-const id = ProviderID.make("google-vertex")
 
 export type OAuthOptions =
   | { readonly accessToken?: string; readonly auth?: never }
@@ -37,18 +35,12 @@ export const host = (location: string) => {
 
 export const requireProject = (value: string | undefined) => {
   if (value) return value
-  throw new ProviderConfigurationError({
-    provider: id,
-    message: "Google Vertex requires a project when baseURL is not configured",
-  })
+  throw new Error("Google Vertex requires a project when baseURL is not configured")
 }
 
 export const apiKey = (input: ApiKeyOptions) => {
   if (input.apiKey !== undefined && (input.accessToken !== undefined || input.auth !== undefined))
-    throw new ProviderConfigurationError({
-      provider: id,
-      message: "Google Vertex apiKey cannot be combined with accessToken or auth",
-    })
+    throw new Error("Google Vertex apiKey cannot be combined with accessToken or auth")
   if (input.accessToken !== undefined || input.auth !== undefined) return undefined
   return input.apiKey ?? process.env.GOOGLE_VERTEX_API_KEY
 }
@@ -76,10 +68,7 @@ const adc = (project?: string) => {
 
 export const oauth = (input: OAuthOptions, project?: string) => {
   if (input.accessToken !== undefined && input.auth !== undefined)
-    throw new ProviderConfigurationError({
-      provider: id,
-      message: "Google Vertex accessToken cannot be combined with auth",
-    })
+    throw new Error("Google Vertex accessToken cannot be combined with auth")
   if (input.auth) return input.auth
   if (input.accessToken !== undefined) return Auth.bearer(input.accessToken)
   return adc(project)

packages/ai/src/providers/google-vertex.ts

@@ -6,7 +6,7 @@ import { Auth } from "../route/auth.js"
 import { Route, type RouteDefaultsInput } from "../route/client.js"
 import { Endpoint } from "../route/endpoint.js"
 import { Framing } from "../route/framing.js"
-import { ProviderConfigurationError, ProviderID, type LLMRequest, type ModelID } from "../schema/index.js"
+import { ProviderID, type LLMRequest, type ModelID } from "../schema/index.js"
 import { GoogleVertexShared } from "./google-vertex-shared.js"
 
 export interface GeminiOptionsInput extends Gemini.OptionsInput {
@@ -93,10 +93,7 @@ const configuredRoute = (input: Config, modelID: string | ModelID) => {
   const apiKey = GoogleVertexShared.apiKey(input)
   const endpointModel = String(modelID).startsWith("endpoints/")
   if (apiKey !== undefined && endpointModel)
-    throw new ProviderConfigurationError({
-      provider: id,
-      message: "Google Vertex tuned models do not support Express Mode API keys",
-    })
+    throw new Error("Google Vertex tuned models do not support Express Mode API keys")
   const location = GoogleVertexShared.location(inputLocation, "us-central1")
   const project = GoogleVertexShared.project(inputProject)
   const endpoint =
@@ -126,10 +123,7 @@ export const provider = {
 }
 export const model: ProviderPackage.Definition<Settings, GeminiProviderOptionsInput>["model"] = (modelID, settings) => {
   if (settings.apiKey !== undefined && settings.accessToken !== undefined)
-    throw new ProviderConfigurationError({
-      provider: id,
-      message: "Google Vertex apiKey cannot be combined with accessToken or auth",
-    })
+    throw new Error("Google Vertex apiKey cannot be combined with accessToken or auth")
   return configure({
     ...(settings.apiKey === undefined ? { accessToken: settings.accessToken } : { apiKey: settings.apiKey }),
     baseURL: settings.baseURL,

packages/ai/src/providers/index.ts

@@ -1,4 +1,3 @@
-export * as Alibaba from "./alibaba.js"
 export * as Anthropic from "./anthropic.js"
 export * as AnthropicCompatible from "./anthropic-compatible.js"
 export * as AmazonBedrock from "./amazon-bedrock.js"

packages/ai/src/providers/xai.ts

@@ -41,10 +41,6 @@ const responsesRoute = Route.make({
     id: "openai-responses",
     name: "xAI Responses",
     rotateAfterMs: RESPONSES_WEBSOCKET_ROTATE_AFTER_MS,
-    // xAI continues a chain only from stored responses: with `store: false` (the route default) `previous_response_id`
-    // fails with "Response with id=… not found", so those steps are sent in full over the reused connection. It also
-    // rejects `instructions` next to `previous_response_id` and keeps the instructions of the response it continues.
-    continuation: ({ instructions: _instructions, ...request }) => (request.store === false ? undefined : request),
   }),
   defaults: { providerOptions: { store: false, include: ["reasoning.encrypted_content"] } },
 })

packages/ai/src/route/client.ts

@@ -7,7 +7,6 @@ import { HttpTransport } from "./transport/index.js"
 import type { HttpMiddleware, Transport, TransportRuntime, WebSocketChannelExecutor } from "./transport/index.js"
 import type { Protocol } from "./protocol.js"
 import { applyCachePolicy } from "../cache-policy.js"
-import { applyEffortUpdates } from "../effort-updates.js"
 import { normalizeToolHistory } from "../tool-history.js"
 import { sanitizeSurrogates } from "../utils/sanitize.js"
 import * as ProviderShared from "../protocols/shared.js"
@@ -24,7 +23,6 @@ import {
   LanguageModel,
   LLMEvent,
   InvalidProviderOutputError,
-  ProviderConfigurationError,
   ProviderID,
   mergeGenerationOptions,
   mergeHttpOptions,
@@ -56,7 +54,6 @@ export interface Route<
   readonly transport: Transport<Body, Prepared, unknown>
   readonly defaults: RouteDefaults
   readonly body: RouteBody<Body>
-  readonly supportsEffortUpdates?: (request: LLMRequest) => boolean
   readonly with: {
     <Next extends CompactionOperations | undefined>(
       patch: RoutePatch<Body, Prepared> & { readonly compact: Next },
@@ -131,10 +128,7 @@ const makeRouteLanguageModel = <Options extends ProviderOptions, Compact extends
   const provider = route.provider ?? ("provider" in mapped ? mapped.provider : undefined)
   if (!provider) throw new Error(`Route.model(${route.id}) requires a provider`)
   if (!endpointBaseURL(route.endpoint))
-    throw new ProviderConfigurationError({
-      provider: ProviderID.make(provider),
-      message: `Route.model(${route.id}) requires an endpoint baseURL — configure it on the route first`,
-    })
+    throw new Error(`Route.model(${route.id}) requires an endpoint baseURL — configure it on the route first`)
   return LanguageModel.make<Options, Compact>({
     ...mapped,
     provider,
@@ -390,7 +384,6 @@ function makeFromTransport<Body, Prepared, Frame, Event, State>(
       transport: routeInput.transport,
       defaults: routeInput.defaults ?? {},
       body: protocol.body,
-      supportsEffortUpdates: protocol.supportsEffortUpdates,
       with: (patch: RoutePatch<Body, Prepared>) => {
         const { compact, id, provider, providerMetadataKey, auth, transport, endpoint, ...defaults } = patch
         return build({
@@ -561,9 +554,7 @@ const prepareRequest = (request: LLMRequest) => {
     [...new Map(tools.map((tool) => [`${tool.type}:${tool.name}`, tool])).values()].map((tool) =>
       tool.type === "tool" ? tool : { ...tool, tools: dedupe(tool.tools) },
     )
-  const resolved = applyCachePolicy(
-    applyEffortUpdates(LLMRequest.update(sanitized, { tools: dedupe(sanitized.tools) })),
-  )
+  const resolved = applyCachePolicy(LLMRequest.update(sanitized, { tools: dedupe(sanitized.tools) }))
   const headers = resolved.model.route.headers?.({ request: resolved })
   return headers === undefined
     ? resolved

packages/ai/src/route/protocol.ts

@@ -41,8 +41,6 @@ export interface Protocol<Body, Frame, Event, State> {
   readonly body: ProtocolBody<Body>
   /** Response side: streaming state machine. */
   readonly stream: ProtocolStream<Frame, Event, State>
-  /** Whether `body.from` lowers `Message.effort(...)` markers; wrappers around another `body.from` must forward it. */
-  readonly supportsEffortUpdates?: (request: LLMRequest) => boolean
 }
 
 export interface ProtocolBody<Body> {

packages/ai/src/route/transport/websocket.ts

@@ -115,11 +115,8 @@ const waitOpen = (ws: globalThis.WebSocket, input: WebSocketRequest) => {
     }
     const onAbort = () => {
       cleanup()
-      if (ws.readyState === globalThis.WebSocket.CLOSED || ws.readyState === globalThis.WebSocket.CLOSING) return
-      // Node's ws reports an aborted handshake as an error event on the next tick; with no listener left
-      // after cleanup, EventEmitter would throw it as an uncaught exception.
-      ws.addEventListener("error", () => {}, { once: true })
-      ws.close(1000)
+      if (ws.readyState !== globalThis.WebSocket.CLOSED && ws.readyState !== globalThis.WebSocket.CLOSING)
+        ws.close(1000)
     }
     const onOpen = () => {
       cleanup()

packages/ai/src/schema/errors.ts

@@ -50,19 +50,6 @@ export class UnsupportedOperationError extends Schema.TaggedError<UnsupportedOpe
   route: Schema.optional(RouteID),
 }) {}
 
-/**
- * Provider settings that are missing, conflicting, or unsupported, such as
- * Azure without `resourceName` or `baseURL`. Thrown synchronously while a
- * provider facade or package entrypoint configures a model, before any
- * request exists, so it is not an `AIError` reason.
- */
-export class ProviderConfigurationError extends Schema.TaggedError<ProviderConfigurationError>(
-  "AI.Error.ProviderConfiguration",
-)("ProviderConfiguration", {
-  provider: ProviderID,
-  message: Schema.String,
-}) {}
-
 export class NoRouteError extends Schema.TaggedError<NoRouteError>("AI.Error.NoRoute")("NoRoute", {
   ...ReasonFields,
   route: RouteID,

packages/ai/src/schema/messages.ts

@@ -9,7 +9,6 @@ import {
   LanguageModelSchema,
   type LanguageModel,
   ProviderOptions,
-  ReasoningEffort,
 } from "./options.js"
 import { ProviderID } from "./ids.js"
 
@@ -218,14 +217,6 @@ export const CompactionPart = Object.assign(compactionPartSchema, {
     Schema.decodeUnknownSync(compactionPartSchema)({ type: "compaction", ...input }),
 })
 
-/** Reasoning effort changed here, from `previous` to `effort`; `undefined` is the model default. */
-export const EffortPart = Schema.Struct({
-  type: Schema.Literal("effort"),
-  effort: Schema.optional(ReasoningEffort),
-  previous: Schema.optional(ReasoningEffort),
-}).annotate({ identifier: "LLM.Content.Effort" })
-export type EffortPart = Schema.Schema.Type<typeof EffortPart>
-
 export const ContentPart = Schema.Union([
   TextPart,
   MediaPart,
@@ -233,7 +224,6 @@ export const ContentPart = Schema.Union([
   ToolResultPart,
   ReasoningPart,
   CompactionPart,
-  EffortPart,
 ]).pipe(Schema.toTaggedUnion("type"))
 export type ContentPart = Schema.Schema.Type<typeof ContentPart>
 
@@ -275,9 +265,6 @@ export namespace Message {
    */
   export const system = (content: SystemContentInput) => make({ role: "system", content })
 
-  export const effort = (input: { readonly effort?: ReasoningEffort; readonly previous?: ReasoningEffort }) =>
-    make({ role: "system", content: [{ type: "effort", effort: input.effort, previous: input.previous }] })
-
   export const tool = (result: ToolResultPart | Parameters<typeof ToolResultPart.make>[0]) =>
     make({ role: "tool", content: ["type" in result ? result : ToolResultPart.make(result)] })
 }

packages/ai/src/schema/options.ts

@@ -140,14 +140,6 @@ export namespace LanguageModelDefaults {
   }
 }
 
-/** Ordered lowest to highest. */
-export const ReasoningEfforts = ["none", "minimal", "low", "medium", "high", "xhigh", "max"] as const
-export type ReasoningEffort = (typeof ReasoningEfforts)[number] | (string & {})
-export const ReasoningEffort = Schema.declare<ReasoningEffort>(
-  (value): value is ReasoningEffort => typeof value === "string",
-  { title: "ReasoningEffort" },
-)
-
 export const LanguageModelToolSchemaCompatibility = Schema.Literals(["gemini", "moonshot"])
 export type LanguageModelToolSchemaCompatibility = Schema.Schema.Type<typeof LanguageModelToolSchemaCompatibility>
 
@@ -173,8 +165,6 @@ export class LanguageModelCompatibility extends Schema.Class<LanguageModelCompat
   requireSignature: Schema.optional(Schema.Boolean),
   /** Supports Anthropic's thinking-prefix mismatch controls. Overrides model-ID detection. */
   supportsThinkingBlockBinding: Schema.optional(Schema.Boolean),
-  /** Supports per-message effort updates. Overrides model-ID detection. */
-  supportsEffortUpdates: Schema.optional(Schema.Boolean),
 }) {}
 
 export namespace LanguageModelCompatibility {

packages/ai/test/effort-updates.test.ts

@@ -1,439 +0,0 @@
-import { describe, expect, test } from "bun:test"
-import { Effect } from "effect"
-import { LLM, LLMRequest, Message, ToolCallPart } from "../src/index.js"
-import { Auth, LLMClient } from "../src/route.js"
-import { compileRequest } from "../src/route/client.js"
-import { AnthropicMessages } from "../src/protocols/anthropic-messages.js"
-import { OpenAIResponses } from "../src/protocols/openai-responses.js"
-import { Gemini } from "../src/protocols/gemini.js"
-import { GoogleVertexMessages, OpenAI } from "../src/providers.js"
-import { applyCachePolicy } from "../src/cache-policy.js"
-import { applyEffortUpdates } from "../src/effort-updates.js"
-import { it, testEffect } from "./lib/effect.js"
-import { dynamicResponse } from "./lib/http.js"
-import { sseEvents } from "./lib/sse.js"
-
-const anthropic = (id: string, compatibility?: { readonly supportsEffortUpdates?: boolean }) =>
-  AnthropicMessages.route
-    .with({ endpoint: { baseURL: "https://api.anthropic.test/v1/" }, auth: Auth.header("x-api-key", "test") })
-    .model({ id, compatibility })
-
-const openai = (id: string, compatibility?: { readonly supportsEffortUpdates?: boolean }) =>
-  OpenAIResponses.route
-    .with({ endpoint: { baseURL: "https://api.openai.test/v1/" }, auth: Auth.bearer("test") })
-    .model({ id, compatibility })
-
-const opus5 = anthropic("claude-opus-5")
-const astra = openai("gpt-6-astra")
-
-const lowFromHigh = Message.effort({ effort: "low", previous: "high" })
-const conversation = [Message.user("Before."), lowFromHigh, Message.user("After.")]
-
-const systemMessages = (body: AnthropicMessages.AnthropicMessagesBody) =>
-  body.messages.filter((message) => message.role === "system")
-
-const updates = (body: OpenAIResponses.OpenAIResponsesBody) =>
-  body.input.filter((item) => "type" in item && item.type === "configuration_update")
-
-describe("applyEffortUpdates", () => {
-  test("keeps the request identity without markers and for protocols that lower them", () => {
-    const plain = LLM.request({ model: opus5, prompt: "hi" })
-    expect(applyEffortUpdates(plain)).toBe(plain)
-
-    const supported = LLM.request({ model: opus5, messages: conversation, providerOptions: { effort: "low" } })
-    expect(applyEffortUpdates(supported)).toBe(supported)
-  })
-
-  it.effect("compiles markers away for protocols without per-message effort", () =>
-    Effect.gen(function* () {
-      const model = Gemini.route
-        .with({
-          endpoint: { baseURL: "https://generativelanguage.test/v1beta/" },
-          auth: Auth.header("x-goog-api-key", "test"),
-        })
-        .model({ id: "gemini-3.5-flash" })
-      const withMarkers = yield* compileRequest(LLM.request({ model, messages: conversation }))
-      const withoutMarkers = yield* compileRequest(
-        LLM.request({ model, messages: [Message.user("Before."), Message.user("After.")] }),
-      )
-
-      expect(withMarkers.body).toEqual(withoutMarkers.body)
-    }),
-  )
-})
-
-describe("cache policy", () => {
-  it.effect("walks the tail breakpoint back past a trailing effort marker", () =>
-    Effect.gen(function* () {
-      const prepared = yield* compileRequest(
-        LLM.request({
-          model: opus5,
-          messages: [Message.user("first"), Message.assistant("reply"), Message.user("latest"), lowFromHigh],
-          providerOptions: { effort: "low" },
-          cache: "auto",
-        }),
-      )
-
-      expect(prepared.body.messages).toEqual([
-        { role: "user", content: [{ type: "text", text: "first" }] },
-        { role: "assistant", content: [{ type: "text", text: "reply" }] },
-        { role: "user", content: [{ type: "text", text: "latest", cache_control: { type: "ephemeral" } }] },
-        { role: "system", content: [], output_config: { effort: "low" } },
-      ])
-    }),
-  )
-})
-
-describe("Anthropic Messages effort updates", () => {
-  it.effect("lowers markers to per-turn system messages and freezes the top-level effort", () =>
-    Effect.gen(function* () {
-      const prepared = yield* compileRequest(
-        LLM.request({ model: opus5, messages: conversation, providerOptions: { effort: "low" }, cache: "none" }),
-      )
-
-      expect(prepared.body.output_config).toEqual({ effort: "high" })
-      expect(prepared.body.messages).toEqual([
-        { role: "user", content: [{ type: "text", text: "Before." }] },
-        { role: "system", content: [], output_config: { effort: "low" } },
-        { role: "user", content: [{ type: "text", text: "After." }] },
-      ])
-    }),
-  )
-
-  it.effect("omits the frozen effort for the model default and sends `high` for a switch back to it", () =>
-    Effect.gen(function* () {
-      const format = { type: "json_schema" as const, schema: { type: "object" } }
-      const prepared = yield* compileRequest(
-        LLM.request({
-          model: opus5,
-          messages: [
-            Message.user("One."),
-            Message.effort({ effort: "low" }),
-            Message.user("Two."),
-            Message.effort({ previous: "low" }),
-            Message.user("Three."),
-          ],
-          providerOptions: { output_config: { format } },
-          cache: "none",
-        }),
-      )
-
-      expect(prepared.body.output_config).toEqual({ format })
-      expect(systemMessages(prepared.body)).toEqual([
-        { role: "system", content: [], output_config: { effort: "low" } },
-        { role: "system", content: [], output_config: { effort: "high" } },
-      ])
-    }),
-  )
-
-  it.effect("requests the mid-conversation output config beta only when markers are sent", () =>
-    Effect.gen(function* () {
-      for (const [model, betas] of [
-        [opus5, "existing-beta,mid-conversation-output-config-2026-07-01"],
-        [anthropic("claude-sonnet-5"), "existing-beta"],
-      ] as const) {
-        const request = LLM.request({
-          model,
-          messages: conversation,
-          providerOptions: { effort: "low" },
-          http: { headers: { "anthropic-beta": "existing-beta" } },
-        })
-        const compiled = yield* compileRequest(request)
-        const prepared = yield* AnthropicMessages.route.prepareTransport(compiled.body, request)
-        expect(prepared.request.headers["anthropic-beta"]).toBe(betas)
-      }
-    }),
-  )
-
-  it.effect("accepts a marker between a tool call and its result", () =>
-    Effect.gen(function* () {
-      const prepared = yield* compileRequest(
-        LLM.request({
-          model: opus5,
-          messages: [
-            Message.user("Weather?"),
-            Message.assistant([ToolCallPart.make({ id: "call_1", name: "lookup", input: {} })]),
-            lowFromHigh,
-            Message.tool({ id: "call_1", name: "lookup", result: { temp: 72 } }),
-          ],
-          providerOptions: { effort: "low" },
-          cache: "none",
-        }),
-      )
-
-      expect(prepared.body.messages).toEqual([
-        { role: "user", content: [{ type: "text", text: "Weather?" }] },
-        { role: "assistant", content: [{ type: "tool_use", id: "call_1", name: "lookup", input: {} }] },
-        { role: "system", content: [], output_config: { effort: "low" } },
-        { role: "user", content: [{ type: "tool_result", tool_use_id: "call_1", content: '{"temp":72}' }] },
-      ])
-    }),
-  )
-
-  it.effect("falls back to a plain top-level effort when history drifted from the current effort", () =>
-    Effect.gen(function* () {
-      const drifted = yield* compileRequest(
-        LLM.request({ model: opus5, messages: conversation, providerOptions: { effort: "medium" }, cache: "none" }),
-      )
-      const plain = yield* compileRequest(
-        LLM.request({
-          model: opus5,
-          messages: [Message.user("Before."), Message.user("After.")],
-          providerOptions: { effort: "medium" },
-          cache: "none",
-        }),
-      )
-
-      expect(drifted.body).toEqual(plain.body)
-      expect(drifted.body.output_config).toEqual({ effort: "medium" })
-    }),
-  )
-
-  for (const [id, supported] of [
-    ["claude-opus-5", true],
-    ["claude-opus-5-20260901", true],
-    ["anthropic/claude-opus-5", true],
-    ["claude-fable-5-1", true],
-    ["claude-mythos-5-1", true],
-    ["claude-fable-5", false],
-    ["claude-opus-4-8", false],
-    ["claude-sonnet-5", false],
-    ["kimi-k2.5", false],
-  ] as const) {
-    it.effect(`${supported ? "lowers" : "strips"} markers for ${id}`, () =>
-      Effect.gen(function* () {
-        const prepared = yield* compileRequest(
-          LLM.request({ model: anthropic(id), messages: conversation, providerOptions: { effort: "low" } }),
-        )
-
-        expect(systemMessages(prepared.body)).toHaveLength(supported ? 1 : 0)
-        expect(prepared.body.output_config).toEqual({ effort: supported ? "high" : "low" })
-      }),
-    )
-  }
-
-  it.effect("honors the compatibility override in both directions", () =>
-    Effect.gen(function* () {
-      const enabled = yield* compileRequest(
-        LLM.request({
-          model: anthropic("claude-sonnet-5", { supportsEffortUpdates: true }),
-          messages: conversation,
-          providerOptions: { effort: "low" },
-        }),
-      )
-      const disabled = yield* compileRequest(
-        LLM.request({
-          model: anthropic("claude-opus-5", { supportsEffortUpdates: false }),
-          messages: conversation,
-          providerOptions: { effort: "low" },
-        }),
-      )
-
-      expect(systemMessages(enabled.body)).toHaveLength(1)
-      expect(systemMessages(disabled.body)).toHaveLength(0)
-    }),
-  )
-
-  it.effect("strips markers on the Vertex Anthropic route, whose protocol wrapper does not forward support", () =>
-    Effect.gen(function* () {
-      const prepared = yield* compileRequest(
-        LLM.request({
-          model: GoogleVertexMessages.configure({ accessToken: "test", location: "global", project: "test" }).model(
-            "claude-opus-5",
-          ),
-          messages: conversation,
-          providerOptions: { effort: "low" },
-        }),
-      )
-
-      expect(systemMessages(prepared.body)).toHaveLength(0)
-      expect(prepared.body.output_config).toEqual({ effort: "low" })
-    }),
-  )
-})
-
-describe("OpenAI Responses effort updates", () => {
-  it.effect("lowers markers to configuration_update items and freezes reasoning.effort", () =>
-    Effect.gen(function* () {
-      const prepared = yield* compileRequest(
-        LLM.request({ model: astra, messages: conversation, providerOptions: { reasoningEffort: "low" } }),
-      )
-
-      expect(prepared.body.reasoning).toEqual({ effort: "high" })
-      expect(prepared.body.input).toEqual([
-        { role: "user", content: [{ type: "input_text", text: "Before." }] },
-        { type: "configuration_update", reasoning: { effort: "low" } },
-        { role: "user", content: [{ type: "input_text", text: "After." }] },
-      ])
-    }),
-  )
-
-  it.effect("coalesces consecutive updates so the newest wins", () =>
-    Effect.gen(function* () {
-      const prepared = yield* compileRequest(
-        LLM.request({
-          model: astra,
-          messages: [
-            Message.user("Before."),
-            Message.effort({ effort: "low", previous: "medium" }),
-            Message.effort({ effort: "xhigh", previous: "low" }),
-            Message.user("After."),
-          ],
-          providerOptions: { reasoningEffort: "xhigh" },
-        }),
-      )
-
-      expect(prepared.body.reasoning).toEqual({ effort: "medium" })
-      expect(prepared.body.input).toEqual([
-        { role: "user", content: [{ type: "input_text", text: "Before." }] },
-        { type: "configuration_update", reasoning: { effort: "xhigh" } },
-        { role: "user", content: [{ type: "input_text", text: "After." }] },
-      ])
-    }),
-  )
-
-  it.effect("omits reasoning.effort for the model default and sends `medium` for a switch back to it", () =>
-    Effect.gen(function* () {
-      const prepared = yield* compileRequest(
-        LLM.request({
-          model: astra,
-          messages: [
-            Message.user("One."),
-            Message.effort({ effort: "low" }),
-            Message.user("Two."),
-            Message.effort({ previous: "low" }),
-            Message.user("Three."),
-          ],
-        }),
-      )
-
-      expect(prepared.body.reasoning).toBeUndefined()
-      expect(updates(prepared.body)).toEqual([
-        { type: "configuration_update", reasoning: { effort: "low" } },
-        { type: "configuration_update", reasoning: { effort: "medium" } },
-      ])
-    }),
-  )
-
-  it.effect("falls back to a plain top-level effort when history drifted from the current effort", () =>
-    Effect.gen(function* () {
-      const drifted = yield* compileRequest(
-        LLM.request({ model: astra, messages: conversation, providerOptions: { reasoningEffort: "xhigh" } }),
-      )
-      const plain = yield* compileRequest(
-        LLM.request({
-          model: astra,
-          messages: [Message.user("Before."), Message.user("After.")],
-          providerOptions: { reasoningEffort: "xhigh" },
-        }),
-      )
-
-      expect(drifted.body).toEqual(plain.body)
-      expect(drifted.body.reasoning).toEqual({ effort: "xhigh" })
-    }),
-  )
-
-  it.effect("strips markers when automatic context management is enabled", () =>
-    Effect.gen(function* () {
-      const prepared = yield* compileRequest(
-        LLM.request({
-          model: astra,
-          messages: conversation,
-          providerOptions: { reasoningEffort: "low", contextManagement: [{ type: "compaction" }] },
-        }),
-      )
-
-      expect(updates(prepared.body)).toEqual([])
-      expect(prepared.body.reasoning).toEqual({ effort: "low" })
-      expect(prepared.body.context_management).toEqual([{ type: "compaction" }])
-    }),
-  )
-
-  for (const [id, supported] of [
-    ["gpt-6-astra", true],
-    ["openai/gpt-6-astra", true],
-    ["gpt-6-astra-2026-09-01", false],
-    ["gpt-5.6-sol", false],
-  ] as const) {
-    it.effect(`${supported ? "lowers" : "strips"} markers for ${id}`, () =>
-      Effect.gen(function* () {
-        const prepared = yield* compileRequest(
-          LLM.request({ model: openai(id), messages: conversation, providerOptions: { reasoningEffort: "low" } }),
-        )
-
-        expect(updates(prepared.body)).toHaveLength(supported ? 1 : 0)
-        expect(prepared.body.reasoning).toEqual({ effort: supported ? "high" : "low" })
-      }),
-    )
-  }
-
-  it.effect("honors the compatibility override in both directions", () =>
-    Effect.gen(function* () {
-      const enabled = yield* compileRequest(
-        LLM.request({
-          model: openai("gpt-5.5", { supportsEffortUpdates: true }),
-          messages: conversation,
-          providerOptions: { reasoningEffort: "low" },
-        }),
-      )
-      const disabled = yield* compileRequest(
-        LLM.request({
-          model: openai("gpt-6-astra", { supportsEffortUpdates: false }),
-          messages: conversation,
-          providerOptions: { reasoningEffort: "low" },
-        }),
-      )
-
-      expect(updates(enabled.body)).toHaveLength(1)
-      expect(updates(disabled.body)).toHaveLength(0)
-    }),
-  )
-
-  const checkpoint = { type: "compaction", id: "cmp_1", encrypted_content: "opaque" }
-  const compactRequest = LLM.request({
-    model: OpenAI.configure({ apiKey: "fixture" }).responses("gpt-6-astra"),
-    messages: conversation,
-    providerOptions: { reasoningEffort: "low" },
-  })
-
-  testEffect(
-    dynamicResponse(({ text, respond }) =>
-      Effect.sync(() => {
-        const body = JSON.parse(text)
-        expect(body.reasoning).toEqual({ effort: "high" })
-        expect(body.input).toEqual([
-          { role: "user", content: [{ type: "input_text", text: "Before." }] },
-          { type: "configuration_update", reasoning: { effort: "low" } },
-          { role: "user", content: [{ type: "input_text", text: "After." }] },
-          { type: "compaction_trigger" },
-        ])
-        return respond(sseEvents({ type: "response.completed", response: { id: "resp_1", output: [checkpoint] } }), {
-          headers: { "content-type": "text/event-stream" },
-        })
-      }),
-    ),
-  ).effect("keeps configuration updates in the checkpoint body", () =>
-    Effect.gen(function* () {
-      const result = yield* LLMClient.compact(compactRequest, { mechanism: "trigger" })
-      expect(result.checkpoint.encrypted).toBe("opaque")
-    }),
-  )
-
-  testEffect(
-    dynamicResponse(({ request, text, respond }) =>
-      Effect.sync(() => {
-        expect(new URL(request.url).pathname).toEndWith("/responses/compact")
-        expect(JSON.parse(text).input).toEqual([
-          { role: "user", content: [{ type: "input_text", text: "Before." }] },
-          { role: "user", content: [{ type: "input_text", text: "After." }] },
-        ])
-        return respond(JSON.stringify({ object: "response.compaction", output: [checkpoint] }))
-      }),
-    ),
-  ).effect("drops markers from the compaction endpoint body", () =>
-    Effect.gen(function* () {
-      const result = yield* LLMClient.compact(compactRequest, { mechanism: "endpoint" })
-      expect(result.replacement.map((message) => message.content[0]?.type)).toEqual(["compaction"])
-    }),
-  )
-})

packages/ai/test/fixtures/recordings/alibaba-chat/qwen-3-7-plus-streams-thinking-disabled.json

@@ -1,35 +0,0 @@
-{
-  "version": 1,
-  "metadata": {
-    "tags": [
-      "prefix:alibaba-chat",
-      "provider:alibaba",
-      "protocol:alibaba-chat",
-      "region:ap-southeast-1",
-      "thinking",
-      "usage"
-    ],
-    "name": "alibaba-chat/qwen-3-7-plus-streams-thinking-disabled",
-    "recordedAt": "2026-09-08T03:10:42.782Z"
-  },
-  "interactions": [
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1/chat/completions",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.7-plus\",\"messages\":[{\"role\":\"user\",\"content\":\"What is 173 multiplied by 219? Reply with only the final integer.\"}],\"stream\":true,\"stream_options\":{\"include_usage\":true},\"max_completion_tokens\":4096,\"enable_thinking\":false}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream;charset=utf-8"
-        },
-        "body": "data: {\"model\":\"qwen3.7-plus\",\"id\":\"chatcmpl-047bcb67-b193-9a4f-9d77-ea0325be6d7c\",\"created\":1788837041,\"object\":\"chat.completion.chunk\",\"usage\":null,\"choices\":[{\"logprobs\":null,\"index\":0,\"delta\":{\"content\":\"\",\"role\":\"assistant\"},\"finish_reason\":null}]}\n\ndata: {\"model\":\"qwen3.7-plus\",\"id\":\"chatcmpl-047bcb67-b193-9a4f-9d77-ea0325be6d7c\",\"choices\":[{\"delta\":{\"content\":\"3\"},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788837041,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.7-plus\",\"id\":\"chatcmpl-047bcb67-b193-9a4f-9d77-ea0325be6d7c\",\"choices\":[{\"delta\":{\"content\":\"7887\"},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788837041,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.7-plus\",\"id\":\"chatcmpl-047bcb67-b193-9a4f-9d77-ea0325be6d7c\",\"choices\":[{\"delta\":{\"content\":\"\"},\"index\":0,\"finish_reason\":\"stop\",\"logprobs\":null}],\"created\":1788837041,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"choices\":[],\"created\":1788837041,\"id\":\"chatcmpl-047bcb67-b193-9a4f-9d77-ea0325be6d7c\",\"model\":\"qwen3.7-plus\",\"object\":\"chat.completion.chunk\",\"usage\":{\"completion_tokens\":5,\"prompt_tokens\":32,\"prompt_tokens_details\":{\"cached_tokens\":0,\"text_tokens\":32},\"total_tokens\":37}}\n\ndata: [DONE]\n\n"
-      }
-    }
-  ]
-}

packages/ai/test/fixtures/recordings/alibaba-chat/qwen-3-8-max-obeys-named-tool-choice.json

@@ -1,35 +0,0 @@
-{
-  "version": 1,
-  "metadata": {
-    "tags": [
-      "prefix:alibaba-chat",
-      "provider:alibaba",
-      "protocol:alibaba-chat",
-      "region:ap-southeast-1",
-      "tool",
-      "tool-choice"
-    ],
-    "name": "alibaba-chat/qwen-3-8-max-obeys-named-tool-choice",
-    "recordedAt": "2026-09-08T03:10:56.576Z"
-  },
-  "interactions": [
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1/chat/completions",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.8-max\",\"messages\":[{\"role\":\"user\",\"content\":\"Find the current weather in Paris.\"}],\"tools\":[{\"type\":\"function\",\"function\":{\"name\":\"get_weather\",\"description\":\"Get weather in a city\",\"parameters\":{\"type\":\"object\",\"properties\":{\"city\":{\"type\":\"string\",\"enum\":[\"Paris\"]}},\"required\":[\"city\"]}}}],\"tool_choice\":{\"type\":\"function\",\"function\":{\"name\":\"get_weather\"}},\"stream\":true,\"stream_options\":{\"include_usage\":true},\"reasoning_effort\":\"none\",\"max_completion_tokens\":4096}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream;charset=utf-8"
-        },
-        "body": "data: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-681dfa5a-ae08-98da-ba0d-4b1d0d364b6b\",\"created\":1788837055,\"object\":\"chat.completion.chunk\",\"usage\":null,\"choices\":[{\"logprobs\":null,\"index\":0,\"delta\":{\"content\":\"\",\"role\":\"assistant\",\"tool_calls\":[{\"index\":0,\"id\":\"call_4eb823cb28d141c8befbb331\",\"type\":\"function\",\"function\":{\"name\":\"get_weather\",\"arguments\":\"\"}}]},\"finish_reason\":null}]}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-681dfa5a-ae08-98da-ba0d-4b1d0d364b6b\",\"choices\":[{\"delta\":{\"content\":\"\",\"tool_calls\":[{\"index\":0,\"id\":\"\",\"type\":\"function\",\"function\":{\"arguments\":\"\"}}]},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788837055,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-681dfa5a-ae08-98da-ba0d-4b1d0d364b6b\",\"choices\":[{\"delta\":{\"content\":\"\",\"tool_calls\":[{\"type\":\"function\",\"index\":0,\"function\":{\"arguments\":\"{\\\"city\\\": \\\"Paris\"}}]},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788837055,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-681dfa5a-ae08-98da-ba0d-4b1d0d364b6b\",\"choices\":[{\"delta\":{\"content\":\"\",\"tool_calls\":[{\"type\":\"function\",\"index\":0,\"function\":{\"arguments\":\"\\\"\"}}]},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788837055,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-681dfa5a-ae08-98da-ba0d-4b1d0d364b6b\",\"choices\":[{\"delta\":{\"content\":\"\",\"tool_calls\":[{\"type\":\"function\",\"index\":0,\"function\":{\"arguments\":\"}\"}}]},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788837055,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-681dfa5a-ae08-98da-ba0d-4b1d0d364b6b\",\"choices\":[{\"delta\":{\"content\":\"\",\"tool_calls\":[{\"type\":\"function\",\"index\":0,\"function\":{\"arguments\":\"\"}}]},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788837055,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-681dfa5a-ae08-98da-ba0d-4b1d0d364b6b\",\"choices\":[{\"delta\":{\"tool_calls\":[{\"function\":{\"arguments\":\"\"},\"index\":0,\"id\":null,\"type\":\"function\"}],\"content\":\"\"},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788837055,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-681dfa5a-ae08-98da-ba0d-4b1d0d364b6b\",\"choices\":[{\"delta\":{},\"index\":0,\"finish_reason\":\"stop\",\"logprobs\":null}],\"created\":1788837055,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"choices\":[],\"created\":1788837055,\"id\":\"chatcmpl-681dfa5a-ae08-98da-ba0d-4b1d0d364b6b\",\"model\":\"qwen3.8-max\",\"object\":\"chat.completion.chunk\",\"usage\":{\"completion_tokens\":19,\"prompt_tokens\":288,\"prompt_tokens_details\":{\"cached_tokens\":0,\"text_tokens\":288},\"total_tokens\":307}}\n\ndata: [DONE]\n\n"
-      }
-    }
-  ]
-}

packages/ai/test/fixtures/recordings/alibaba-chat/qwen-3-8-max-returns-a-json-object.json

@@ -1,34 +0,0 @@
-{
-  "version": 1,
-  "metadata": {
-    "tags": [
-      "prefix:alibaba-chat",
-      "provider:alibaba",
-      "protocol:alibaba-chat",
-      "region:ap-southeast-1",
-      "structured-output"
-    ],
-    "name": "alibaba-chat/qwen-3-8-max-returns-a-json-object",
-    "recordedAt": "2026-09-08T03:11:29.462Z"
-  },
-  "interactions": [
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1/chat/completions",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.8-max\",\"messages\":[{\"role\":\"user\",\"content\":\"Return a JSON object with one key \\\"city\\\" set to the capital city of France.\"}],\"stream\":true,\"stream_options\":{\"include_usage\":true},\"reasoning_effort\":\"none\",\"max_completion_tokens\":1024,\"response_format\":{\"type\":\"json_object\"}}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream;charset=utf-8"
-        },
-        "body": "data: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-de34af77-7b1e-9999-a48f-e564c03d4b6b\",\"created\":1788837088,\"object\":\"chat.completion.chunk\",\"usage\":null,\"choices\":[{\"logprobs\":null,\"index\":0,\"delta\":{\"content\":\"\",\"role\":\"assistant\"},\"finish_reason\":null}]}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-de34af77-7b1e-9999-a48f-e564c03d4b6b\",\"choices\":[{\"delta\":{\"content\":\"{\\\"\"},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788837088,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-de34af77-7b1e-9999-a48f-e564c03d4b6b\",\"choices\":[{\"delta\":{\"content\":\"city\\\":\"},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788837088,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-de34af77-7b1e-9999-a48f-e564c03d4b6b\",\"choices\":[{\"delta\":{\"content\":\" \\\"Paris\\\"}\"},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788837088,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-de34af77-7b1e-9999-a48f-e564c03d4b6b\",\"choices\":[{\"delta\":{\"content\":\"\"},\"index\":0,\"finish_reason\":\"stop\",\"logprobs\":null}],\"created\":1788837088,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"choices\":[],\"created\":1788837088,\"id\":\"chatcmpl-de34af77-7b1e-9999-a48f-e564c03d4b6b\",\"model\":\"qwen3.8-max\",\"object\":\"chat.completion.chunk\",\"usage\":{\"completion_tokens\":6,\"prompt_tokens\":32,\"prompt_tokens_details\":{\"cached_tokens\":0,\"text_tokens\":32},\"total_tokens\":38}}\n\ndata: [DONE]\n\n"
-      }
-    }
-  ]
-}

packages/ai/test/fixtures/recordings/alibaba-chat/qwen-3-8-max-streams-none-effort.json

@@ -1,36 +0,0 @@
-{
-  "version": 1,
-  "metadata": {
-    "tags": [
-      "prefix:alibaba-chat",
-      "provider:alibaba",
-      "protocol:alibaba-chat",
-      "region:ap-southeast-1",
-      "text",
-      "reasoning",
-      "usage"
-    ],
-    "name": "alibaba-chat/qwen-3-8-max-streams-none-effort",
-    "recordedAt": "2026-09-08T03:09:27.584Z"
-  },
-  "interactions": [
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1/chat/completions",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.8-max\",\"messages\":[{\"role\":\"user\",\"content\":\"What is 173 multiplied by 219? Reply with only the final integer.\"}],\"stream\":true,\"stream_options\":{\"include_usage\":true},\"reasoning_effort\":\"none\",\"max_completion_tokens\":4096}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream;charset=utf-8"
-        },
-        "body": "data: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-d8fbc8fe-4d5a-9dc9-84c6-ed0c7b9dd3bb\",\"created\":1788836967,\"object\":\"chat.completion.chunk\",\"usage\":null,\"choices\":[{\"logprobs\":null,\"index\":0,\"delta\":{\"content\":\"\",\"role\":\"assistant\"},\"finish_reason\":null}]}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-d8fbc8fe-4d5a-9dc9-84c6-ed0c7b9dd3bb\",\"choices\":[{\"delta\":{\"content\":\"3\"},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788836967,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-d8fbc8fe-4d5a-9dc9-84c6-ed0c7b9dd3bb\",\"choices\":[{\"delta\":{\"content\":\"78\"},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788836967,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-d8fbc8fe-4d5a-9dc9-84c6-ed0c7b9dd3bb\",\"choices\":[{\"delta\":{\"content\":\"87\"},\"index\":0,\"finish_reason\":null,\"logprobs\":null}],\"created\":1788836967,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"model\":\"qwen3.8-max\",\"id\":\"chatcmpl-d8fbc8fe-4d5a-9dc9-84c6-ed0c7b9dd3bb\",\"choices\":[{\"delta\":{\"content\":\"\"},\"index\":0,\"finish_reason\":\"stop\",\"logprobs\":null}],\"created\":1788836967,\"object\":\"chat.completion.chunk\",\"usage\":null}\n\ndata: {\"choices\":[],\"created\":1788836967,\"id\":\"chatcmpl-d8fbc8fe-4d5a-9dc9-84c6-ed0c7b9dd3bb\",\"model\":\"qwen3.8-max\",\"object\":\"chat.completion.chunk\",\"usage\":{\"completion_tokens\":5,\"prompt_tokens\":32,\"prompt_tokens_details\":{\"cached_tokens\":0,\"text_tokens\":32},\"total_tokens\":37}}\n\ndata: [DONE]\n\n"
-      }
-    }
-  ]
-}

packages/ai/test/fixtures/recordings/alibaba-messages/qwen-3-7-plus-streams-thinking-disabled.json

@@ -1,35 +0,0 @@
-{
-  "version": 1,
-  "metadata": {
-    "tags": [
-      "prefix:alibaba-messages",
-      "provider:alibaba",
-      "protocol:alibaba-messages",
-      "region:ap-southeast-1",
-      "thinking",
-      "usage"
-    ],
-    "name": "alibaba-messages/qwen-3-7-plus-streams-thinking-disabled",
-    "recordedAt": "2026-09-08T03:10:57.819Z"
-  },
-  "interactions": [
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/apps/anthropic/v1/messages",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.7-plus\",\"messages\":[{\"role\":\"user\",\"content\":[{\"type\":\"text\",\"text\":\"What is 173 multiplied by 219? Reply with only the final integer.\"}]}],\"stream\":true,\"max_tokens\":4096,\"thinking\":{\"type\":\"disabled\"}}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream"
-        },
-        "body": "event:ping\ndata:{\"type\":\"ping\"}\n\nevent:message_start\ndata:{\"message\":{\"model\":\"qwen3.7-plus\",\"id\":\"msg_c4d58b4f-a61d-9d0c-a9e4-cb45d34b1120\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[],\"usage\":{\"input_tokens\":20,\"output_tokens\":0}},\"type\":\"message_start\"}\n\nevent:content_block_start\ndata:{\"type\":\"content_block_start\",\"content_block\":{\"type\":\"text\",\"text\":\"\"},\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"text_delta\",\"text\":\"3\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"text_delta\",\"text\":\"7887\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_stop\ndata:{\"type\":\"content_block_stop\",\"index\":0}\n\nevent:message_delta\ndata:{\"delta\":{\"stop_reason\":\"end_turn\"},\"type\":\"message_delta\",\"usage\":{\"output_tokens\":5,\"cache_creation_input_tokens\":0,\"input_tokens\":32,\"cache_read_input_tokens\":0,\"prompt_tokens_details\":{\"cached_tokens\":0}}}\n\nevent:message_stop\ndata:{\"type\":\"message_stop\"}\n\n"
-      }
-    }
-  ]
-}

packages/ai/test/fixtures/recordings/alibaba-messages/qwen-3-8-max-follows-a-json-schema.json

@@ -1,34 +0,0 @@
-{
-  "version": 1,
-  "metadata": {
-    "tags": [
-      "prefix:alibaba-messages",
-      "provider:alibaba",
-      "protocol:alibaba-messages",
-      "region:ap-southeast-1",
-      "structured-output"
-    ],
-    "name": "alibaba-messages/qwen-3-8-max-follows-a-json-schema",
-    "recordedAt": "2026-09-08T03:11:30.530Z"
-  },
-  "interactions": [
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/apps/anthropic/v1/messages",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.8-max\",\"messages\":[{\"role\":\"user\",\"content\":[{\"type\":\"text\",\"text\":\"Return a JSON object with one key \\\"city\\\" set to the capital city of France.\"}]}],\"stream\":true,\"max_tokens\":1024,\"thinking\":{\"type\":\"disabled\"},\"output_config\":{\"format\":{\"type\":\"json_schema\",\"schema\":{\"type\":\"object\",\"properties\":{\"city\":{\"type\":\"string\"}},\"required\":[\"city\"],\"additionalProperties\":false}}}}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream"
-        },
-        "body": "event:ping\ndata:{\"type\":\"ping\"}\n\nevent:message_start\ndata:{\"message\":{\"model\":\"qwen3.8-max\",\"id\":\"msg_16e067c1-984b-94d3-9abb-11792d794271\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[],\"usage\":{\"input_tokens\":18,\"output_tokens\":0}},\"type\":\"message_start\"}\n\nevent:content_block_start\ndata:{\"type\":\"content_block_start\",\"content_block\":{\"type\":\"text\",\"text\":\"\"},\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"text_delta\",\"text\":\"{\\\"\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"text_delta\",\"text\":\"city\\\":\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"text_delta\",\"text\":\" \\\"Paris\\\"}\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_stop\ndata:{\"type\":\"content_block_stop\",\"index\":0}\n\nevent:message_delta\ndata:{\"delta\":{\"stop_reason\":\"end_turn\"},\"type\":\"message_delta\",\"usage\":{\"output_tokens\":6,\"cache_creation_input_tokens\":0,\"input_tokens\":32,\"cache_read_input_tokens\":0,\"prompt_tokens_details\":{\"cached_tokens\":0}}}\n\nevent:message_stop\ndata:{\"type\":\"message_stop\"}\n\n"
-      }
-    }
-  ]
-}

packages/ai/test/fixtures/recordings/alibaba-messages/qwen-3-8-max-obeys-named-tool-choice.json

@@ -1,35 +0,0 @@
-{
-  "version": 1,
-  "metadata": {
-    "tags": [
-      "prefix:alibaba-messages",
-      "provider:alibaba",
-      "protocol:alibaba-messages",
-      "region:ap-southeast-1",
-      "tool",
-      "tool-choice"
-    ],
-    "name": "alibaba-messages/qwen-3-8-max-obeys-named-tool-choice",
-    "recordedAt": "2026-09-08T03:11:06.590Z"
-  },
-  "interactions": [
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/apps/anthropic/v1/messages",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.8-max\",\"messages\":[{\"role\":\"user\",\"content\":[{\"type\":\"text\",\"text\":\"Find the current weather in Paris.\"}]}],\"tools\":[{\"name\":\"get_weather\",\"description\":\"Get weather in a city\",\"input_schema\":{\"type\":\"object\",\"properties\":{\"city\":{\"type\":\"string\",\"enum\":[\"Paris\"]}},\"required\":[\"city\"]}}],\"tool_choice\":{\"type\":\"tool\",\"name\":\"get_weather\"},\"stream\":true,\"max_tokens\":4096,\"thinking\":{\"type\":\"disabled\"}}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream"
-        },
-        "body": "event:ping\ndata:{\"type\":\"ping\"}\n\nevent:message_start\ndata:{\"message\":{\"model\":\"qwen3.8-max\",\"id\":\"msg_0e8f1abf-f2bc-9a86-a6a7-14804ac17eac\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[],\"usage\":{\"input_tokens\":45,\"output_tokens\":0}},\"type\":\"message_start\"}\n\nevent:content_block_start\ndata:{\"type\":\"content_block_start\",\"content_block\":{\"name\":\"get_weather\",\"input\":{},\"id\":\"toolu_cf9cab33261f4709ae096d8a\",\"type\":\"tool_use\"},\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"partial_json\":\"\",\"type\":\"input_json_delta\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"partial_json\":\"{\\\"city\\\": \\\"Paris\",\"type\":\"input_json_delta\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"partial_json\":\"\\\"\",\"type\":\"input_json_delta\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"partial_json\":\"}\",\"type\":\"input_json_delta\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_stop\ndata:{\"type\":\"content_block_stop\",\"index\":0}\n\nevent:message_delta\ndata:{\"delta\":{\"stop_reason\":\"end_turn\"},\"type\":\"message_delta\",\"usage\":{\"output_tokens\":19,\"cache_creation_input_tokens\":0,\"input_tokens\":288,\"cache_read_input_tokens\":0,\"prompt_tokens_details\":{\"cached_tokens\":0}}}\n\nevent:message_stop\ndata:{\"type\":\"message_stop\"}\n\n"
-      }
-    }
-  ]
-}

packages/ai/test/fixtures/recordings/alibaba-messages/qwen-3-8-max-streams-max-effort.json

@@ -1,36 +0,0 @@
-{
-  "version": 1,
-  "metadata": {
-    "tags": [
-      "prefix:alibaba-messages",
-      "provider:alibaba",
-      "protocol:alibaba-messages",
-      "region:ap-southeast-1",
-      "text",
-      "reasoning",
-      "usage"
-    ],
-    "name": "alibaba-messages/qwen-3-8-max-streams-max-effort",
-    "recordedAt": "2026-09-08T03:12:36.479Z"
-  },
-  "interactions": [
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/apps/anthropic/v1/messages",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.8-max\",\"messages\":[{\"role\":\"user\",\"content\":[{\"type\":\"text\",\"text\":\"What is 173 multiplied by 219? Reply with only the final integer.\"}]}],\"stream\":true,\"max_tokens\":4096,\"output_config\":{\"effort\":\"max\"}}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream"
-        },
-        "body": "event:ping\ndata:{\"type\":\"ping\"}\n\nevent:message_start\ndata:{\"message\":{\"model\":\"qwen3.8-max\",\"id\":\"msg_7952446b-0176-9526-ad7d-b1f0184bc106\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[],\"usage\":{\"input_tokens\":20,\"output_tokens\":0}},\"type\":\"message_start\"}\n\nevent:content_block_start\ndata:{\"type\":\"content_block_start\",\"content_block\":{\"type\":\"thinking\",\"signature\":\"\",\"thinking\":\"\"},\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"We\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\" need answer user\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"'s simple multiplication with\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\" only final integer.\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\" Need compute 1\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"73*2\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"19. \"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"173*\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"200=\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"3460\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"0; 1\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"73*1\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"9=32\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"87 (\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"173*\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"20=3\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"460-\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"173=\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"3287\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"); sum=3\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"7887\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\". Final only integer\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\".\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"signature_delta\",\"signature\":\"\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_stop\ndata:{\"type\":\"content_block_stop\",\"index\":0}\n\nevent:content_block_start\ndata:{\"type\":\"content_block_start\",\"content_block\":{\"type\":\"text\",\"text\":\"\"},\"index\":1}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"text_delta\",\"text\":\"37\"},\"type\":\"content_block_delta\",\"index\":1}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"text_delta\",\"text\":\"887\"},\"type\":\"content_block_delta\",\"index\":1}\n\nevent:content_block_stop\ndata:{\"type\":\"content_block_stop\",\"index\":1}\n\nevent:message_delta\ndata:{\"delta\":{\"stop_reason\":\"end_turn\"},\"type\":\"message_delta\",\"usage\":{\"output_tokens\":92,\"cache_creation_input_tokens\":0,\"input_tokens\":81,\"cache_read_input_tokens\":0,\"prompt_tokens_details\":{\"cached_tokens\":0}}}\n\nevent:message_stop\ndata:{\"type\":\"message_stop\"}\n\n"
-      }
-    }
-  ]
-}

packages/ai/test/fixtures/recordings/alibaba-messages/qwen-3-8-max-streams-xhigh-effort.json

@@ -1,36 +0,0 @@
-{
-  "version": 1,
-  "metadata": {
-    "tags": [
-      "prefix:alibaba-messages",
-      "provider:alibaba",
-      "protocol:alibaba-messages",
-      "region:ap-southeast-1",
-      "text",
-      "reasoning",
-      "usage"
-    ],
-    "name": "alibaba-messages/qwen-3-8-max-streams-xhigh-effort",
-    "recordedAt": "2026-09-08T03:10:04.633Z"
-  },
-  "interactions": [
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/apps/anthropic/v1/messages",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.8-max\",\"messages\":[{\"role\":\"user\",\"content\":[{\"type\":\"text\",\"text\":\"What is 173 multiplied by 219? Reply with only the final integer.\"}]}],\"stream\":true,\"max_tokens\":4096,\"output_config\":{\"effort\":\"xhigh\"}}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream"
-        },
-        "body": "event:ping\ndata:{\"type\":\"ping\"}\n\nevent:message_start\ndata:{\"message\":{\"model\":\"qwen3.8-max\",\"id\":\"msg_a57b8563-71fb-9248-a31e-c49d6ba38717\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[],\"usage\":{\"input_tokens\":20,\"output_tokens\":0}},\"type\":\"message_start\"}\n\nevent:content_block_start\ndata:{\"type\":\"content_block_start\",\"content_block\":{\"type\":\"thinking\",\"signature\":\"\",\"thinking\":\"\"},\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"We\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\" need answer simple multiplication\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\". We\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\" already call\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\". Need compute \"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"173*\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"219.\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\" 173\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"*200\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"=346\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"00; *\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"19=3\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"287;\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\" sum 37\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"88\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\"7. Final only\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\" integer. Ensure\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"thinking_delta\",\"thinking\":\" no extra.\\n\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"signature_delta\",\"signature\":\"\"},\"type\":\"content_block_delta\",\"index\":0}\n\nevent:content_block_stop\ndata:{\"type\":\"content_block_stop\",\"index\":0}\n\nevent:content_block_start\ndata:{\"type\":\"content_block_start\",\"content_block\":{\"type\":\"text\",\"text\":\"\"},\"index\":1}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"text_delta\",\"text\":\"378\"},\"type\":\"content_block_delta\",\"index\":1}\n\nevent:content_block_delta\ndata:{\"delta\":{\"type\":\"text_delta\",\"text\":\"87\"},\"type\":\"content_block_delta\",\"index\":1}\n\nevent:content_block_stop\ndata:{\"type\":\"content_block_stop\",\"index\":1}\n\nevent:message_delta\ndata:{\"delta\":{\"stop_reason\":\"end_turn\"},\"type\":\"message_delta\",\"usage\":{\"output_tokens\":69,\"cache_creation_input_tokens\":0,\"input_tokens\":81,\"cache_read_input_tokens\":0,\"prompt_tokens_details\":{\"cached_tokens\":0}}}\n\nevent:message_stop\ndata:{\"type\":\"message_stop\"}\n\n"
-      }
-    }
-  ]
-}

packages/ai/test/fixtures/recordings/alibaba-responses/qwen-3-7-plus-streams-thinking-disabled.json

@@ -1,35 +0,0 @@
-{
-  "version": 1,
-  "metadata": {
-    "tags": [
-      "prefix:alibaba-responses",
-      "provider:alibaba",
-      "protocol:alibaba-responses",
-      "region:ap-southeast-1",
-      "thinking",
-      "usage"
-    ],
-    "name": "alibaba-responses/qwen-3-7-plus-streams-thinking-disabled",
-    "recordedAt": "2026-09-08T03:11:07.495Z"
-  },
-  "interactions": [
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1/responses",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.7-plus\",\"input\":[{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"What is 173 multiplied by 219? Reply with only the final integer.\"}]}],\"max_output_tokens\":4096,\"enable_thinking\":false,\"stream\":true}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream;charset=UTF-8"
-        },
-        "body": "id:1\nevent:response.created\n:HTTP_STATUS/200\ndata:{\"sequence_number\":0,\"type\":\"response.created\",\"response\":{\"top_logprobs\":0,\"metadata\":{},\"presence_penalty\":0.0,\"created_at\":1788837067,\"store\":true,\"tools\":[],\"output\":[],\"top_p\":1.0,\"completed_at\":1788837067,\"frequency_penalty\":0.0,\"parallel_tool_calls\":true,\"background\":false,\"temperature\":1.0,\"tool_choice\":\"auto\",\"model\":\"qwen3.7-plus\",\"service_tier\":\"default\",\"id\":\"resp_493607bf-baef-9ced-9e36-2e2a583b6177\",\"max_output_tokens\":4096,\"object\":\"response\",\"status\":\"queued\"}}\n\nid:2\nevent:response.in_progress\n:HTTP_STATUS/200\ndata:{\"sequence_number\":1,\"type\":\"response.in_progress\",\"response\":{\"top_logprobs\":0,\"metadata\":{},\"presence_penalty\":0.0,\"created_at\":1788837067,\"store\":true,\"tools\":[],\"output\":[],\"top_p\":1.0,\"completed_at\":1788837067,\"frequency_penalty\":0.0,\"parallel_tool_calls\":true,\"background\":false,\"temperature\":1.0,\"tool_choice\":\"auto\",\"model\":\"qwen3.7-plus\",\"service_tier\":\"default\",\"id\":\"resp_493607bf-baef-9ced-9e36-2e2a583b6177\",\"max_output_tokens\":4096,\"object\":\"response\",\"status\":\"in_progress\"}}\n\nid:3\nevent:response.output_item.added\n:HTTP_STATUS/200\ndata:{\"sequence_number\":2,\"item\":{\"id\":\"msg_509c6dec-e527-41b5-a9c9-7f02fecb7a3f\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[],\"status\":\"in_progress\"},\"output_index\":0,\"type\":\"response.output_item.added\"}\n\nid:4\nevent:response.content_part.added\n:HTTP_STATUS/200\ndata:{\"sequence_number\":3,\"output_index\":0,\"type\":\"response.content_part.added\",\"content_index\":0,\"item_id\":\"msg_509c6dec-e527-41b5-a9c9-7f02fecb7a3f\",\"part\":{\"type\":\"output_text\",\"annotations\":[],\"text\":\"\"}}\n\nid:5\nevent:response.output_text.delta\n:HTTP_STATUS/200\ndata:{\"sequence_number\":4,\"content_index\":0,\"item_id\":\"msg_509c6dec-e527-41b5-a9c9-7f02fecb7a3f\",\"delta\":\"3\",\"output_index\":0,\"type\":\"response.output_text.delta\",\"logprobs\":[]}\n\nid:6\nevent:response.output_text.delta\n:HTTP_STATUS/200\ndata:{\"sequence_number\":5,\"content_index\":0,\"item_id\":\"msg_509c6dec-e527-41b5-a9c9-7f02fecb7a3f\",\"delta\":\"7887\",\"output_index\":0,\"type\":\"response.output_text.delta\",\"logprobs\":[]}\n\nid:7\nevent:response.output_text.delta\n:HTTP_STATUS/200\ndata:{\"sequence_number\":6,\"content_index\":0,\"item_id\":\"msg_509c6dec-e527-41b5-a9c9-7f02fecb7a3f\",\"delta\":\"\",\"output_index\":0,\"type\":\"response.output_text.delta\",\"logprobs\":[]}\n\nid:8\nevent:response.output_text.done\n:HTTP_STATUS/200\ndata:{\"sequence_number\":7,\"content_index\":0,\"item_id\":\"msg_509c6dec-e527-41b5-a9c9-7f02fecb7a3f\",\"text\":\"37887\",\"output_index\":0,\"type\":\"response.output_text.done\",\"logprobs\":[]}\n\nid:9\nevent:response.content_part.done\n:HTTP_STATUS/200\ndata:{\"sequence_number\":8,\"output_index\":0,\"type\":\"response.content_part.done\",\"content_index\":0,\"item_id\":\"msg_509c6dec-e527-41b5-a9c9-7f02fecb7a3f\",\"part\":{\"type\":\"output_text\",\"annotations\":[],\"text\":\"37887\"}}\n\nid:10\nevent:response.output_item.done\n:HTTP_STATUS/200\ndata:{\"sequence_number\":9,\"item\":{\"id\":\"msg_509c6dec-e527-41b5-a9c9-7f02fecb7a3f\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[{\"type\":\"output_text\",\"annotations\":[],\"text\":\"37887\"}],\"status\":\"completed\"},\"output_index\":0,\"type\":\"response.output_item.done\"}\n\nid:11\nevent:response.completed\n:HTTP_STATUS/200\ndata:{\"sequence_number\":10,\"type\":\"response.completed\",\"response\":{\"top_logprobs\":0,\"metadata\":{},\"presence_penalty\":0.0,\"usage\":{\"total_tokens\":73,\"input_tokens_details\":{\"cached_tokens\":0},\"output_tokens\":5,\"input_tokens\":68,\"output_tokens_details\":{\"reasoning_tokens\":0},\"x_details\":[{\"total_tokens\":73,\"x_billing_type\":\"response_api\",\"output_tokens\":5,\"input_tokens\":68,\"prompt_tokens_details\":{\"cached_tokens\":0}}]},\"created_at\":1788837067,\"store\":true,\"tools\":[],\"output\":[{\"id\":\"msg_509c6dec-e527-41b5-a9c9-7f02fecb7a3f\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[{\"type\":\"output_text\",\"annotations\":[],\"text\":\"37887\"}],\"status\":\"completed\"}],\"top_p\":1.0,\"completed_at\":1788837067,\"frequency_penalty\":0.0,\"parallel_tool_calls\":true,\"background\":false,\"temperature\":1.0,\"tool_choice\":\"auto\",\"model\":\"qwen3.7-plus\",\"service_tier\":\"default\",\"id\":\"resp_493607bf-baef-9ced-9e36-2e2a583b6177\",\"max_output_tokens\":4096,\"object\":\"response\",\"status\":\"completed\"}}\n\n"
-      }
-    }
-  ]
-}

packages/ai/test/fixtures/recordings/alibaba-responses/qwen-3-8-max-continues-a-stored-response.json

@@ -1,53 +0,0 @@
-{
-  "version": 1,
-  "metadata": {
-    "tags": [
-      "prefix:alibaba-responses",
-      "provider:alibaba",
-      "protocol:alibaba-responses",
-      "region:ap-southeast-1",
-      "continuation",
-      "storage"
-    ],
-    "name": "alibaba-responses/qwen-3-8-max-continues-a-stored-response",
-    "recordedAt": "2026-09-08T03:12:42.429Z"
-  },
-  "interactions": [
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1/responses",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.8-max\",\"input\":[{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"Remember the password word apricot. Reply OK.\"}]}],\"store\":true,\"reasoning\":{\"effort\":\"none\"},\"max_output_tokens\":1024,\"stream\":true}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream;charset=UTF-8"
-        },
-        "body": "id:1\nevent:response.created\n:HTTP_STATUS/200\ndata:{\"sequence_number\":0,\"type\":\"response.created\",\"response\":{\"top_logprobs\":0,\"metadata\":{},\"presence_penalty\":0.0,\"reasoning\":{\"effort\":\"none\"},\"created_at\":1788837161,\"store\":true,\"tools\":[],\"output\":[],\"top_p\":1.0,\"completed_at\":1788837161,\"frequency_penalty\":0.0,\"parallel_tool_calls\":true,\"background\":false,\"temperature\":1.0,\"tool_choice\":\"auto\",\"model\":\"qwen3.8-max\",\"service_tier\":\"default\",\"id\":\"resp_97f8cd28-c7ab-9d51-9c14-4fc531decdc8\",\"max_output_tokens\":1024,\"object\":\"response\",\"status\":\"queued\"}}\n\nid:2\nevent:response.in_progress\n:HTTP_STATUS/200\ndata:{\"sequence_number\":1,\"type\":\"response.in_progress\",\"response\":{\"top_logprobs\":0,\"metadata\":{},\"presence_penalty\":0.0,\"reasoning\":{\"effort\":\"none\"},\"created_at\":1788837161,\"store\":true,\"tools\":[],\"output\":[],\"top_p\":1.0,\"completed_at\":1788837161,\"frequency_penalty\":0.0,\"parallel_tool_calls\":true,\"background\":false,\"temperature\":1.0,\"tool_choice\":\"auto\",\"model\":\"qwen3.8-max\",\"service_tier\":\"default\",\"id\":\"resp_97f8cd28-c7ab-9d51-9c14-4fc531decdc8\",\"max_output_tokens\":1024,\"object\":\"response\",\"status\":\"in_progress\"}}\n\nid:3\nevent:response.output_item.added\n:HTTP_STATUS/200\ndata:{\"sequence_number\":2,\"item\":{\"id\":\"msg_a908ad8e-6220-444b-880c-02aecbaf8fac\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[],\"status\":\"in_progress\"},\"output_index\":0,\"type\":\"response.output_item.added\"}\n\nid:4\nevent:response.content_part.added\n:HTTP_STATUS/200\ndata:{\"sequence_number\":3,\"output_index\":0,\"type\":\"response.content_part.added\",\"content_index\":0,\"item_id\":\"msg_a908ad8e-6220-444b-880c-02aecbaf8fac\",\"part\":{\"type\":\"output_text\",\"annotations\":[],\"text\":\"\"}}\n\nid:5\nevent:response.output_text.delta\n:HTTP_STATUS/200\ndata:{\"sequence_number\":4,\"content_index\":0,\"item_id\":\"msg_a908ad8e-6220-444b-880c-02aecbaf8fac\",\"delta\":\"OK\",\"output_index\":0,\"type\":\"response.output_text.delta\",\"logprobs\":[]}\n\nid:6\nevent:response.output_text.delta\n:HTTP_STATUS/200\ndata:{\"sequence_number\":5,\"content_index\":0,\"item_id\":\"msg_a908ad8e-6220-444b-880c-02aecbaf8fac\",\"delta\":\".\",\"output_index\":0,\"type\":\"response.output_text.delta\",\"logprobs\":[]}\n\nid:7\nevent:response.output_text.delta\n:HTTP_STATUS/200\ndata:{\"sequence_number\":6,\"content_index\":0,\"item_id\":\"msg_a908ad8e-6220-444b-880c-02aecbaf8fac\",\"delta\":\"\",\"output_index\":0,\"type\":\"response.output_text.delta\",\"logprobs\":[]}\n\nid:8\nevent:response.output_text.done\n:HTTP_STATUS/200\ndata:{\"sequence_number\":7,\"content_index\":0,\"item_id\":\"msg_a908ad8e-6220-444b-880c-02aecbaf8fac\",\"text\":\"OK.\",\"output_index\":0,\"type\":\"response.output_text.done\",\"logprobs\":[]}\n\nid:9\nevent:response.content_part.done\n:HTTP_STATUS/200\ndata:{\"sequence_number\":8,\"output_index\":0,\"type\":\"response.content_part.done\",\"content_index\":0,\"item_id\":\"msg_a908ad8e-6220-444b-880c-02aecbaf8fac\",\"part\":{\"type\":\"output_text\",\"annotations\":[],\"text\":\"OK.\"}}\n\nid:10\nevent:response.output_item.done\n:HTTP_STATUS/200\ndata:{\"sequence_number\":9,\"item\":{\"id\":\"msg_a908ad8e-6220-444b-880c-02aecbaf8fac\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[{\"type\":\"output_text\",\"annotations\":[],\"text\":\"OK.\"}],\"status\":\"completed\"},\"output_index\":0,\"type\":\"response.output_item.done\"}\n\nid:11\nevent:response.completed\n:HTTP_STATUS/200\ndata:{\"sequence_number\":10,\"type\":\"response.completed\",\"response\":{\"top_logprobs\":0,\"metadata\":{},\"presence_penalty\":0.0,\"reasoning\":{\"effort\":\"none\"},\"usage\":{\"total_tokens\":60,\"input_tokens_details\":{\"cached_tokens\":0},\"output_tokens\":2,\"input_tokens\":58,\"output_tokens_details\":{\"reasoning_tokens\":0},\"x_details\":[{\"total_tokens\":60,\"x_billing_type\":\"response_api\",\"output_tokens\":2,\"input_tokens\":58,\"prompt_tokens_details\":{\"cached_tokens\":0}}]},\"created_at\":1788837161,\"store\":true,\"tools\":[],\"output\":[{\"id\":\"msg_a908ad8e-6220-444b-880c-02aecbaf8fac\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[{\"type\":\"output_text\",\"annotations\":[],\"text\":\"OK.\"}],\"status\":\"completed\"}],\"top_p\":1.0,\"completed_at\":1788837161,\"frequency_penalty\":0.0,\"parallel_tool_calls\":true,\"background\":false,\"temperature\":1.0,\"tool_choice\":\"auto\",\"model\":\"qwen3.8-max\",\"service_tier\":\"default\",\"id\":\"resp_97f8cd28-c7ab-9d51-9c14-4fc531decdc8\",\"max_output_tokens\":1024,\"object\":\"response\",\"status\":\"completed\"}}\n\n"
-      }
-    },
-    {
-      "transport": "http",
-      "request": {
-        "method": "POST",
-        "url": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1/responses",
-        "headers": {
-          "content-type": "application/json"
-        },
-        "body": "{\"model\":\"qwen3.8-max\",\"input\":[{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"What word did I ask you to remember? Reply with only the word.\"}]}],\"store\":true,\"reasoning\":{\"effort\":\"none\"},\"max_output_tokens\":1024,\"previous_response_id\":\"resp_97f8cd28-c7ab-9d51-9c14-4fc531decdc8\",\"stream\":true}"
-      },
-      "response": {
-        "status": 200,
-        "headers": {
-          "content-type": "text/event-stream;charset=UTF-8"
-        },
-        "body": "id:1\nevent:response.created\n:HTTP_STATUS/200\ndata:{\"sequence_number\":0,\"type\":\"response.created\",\"response\":{\"top_logprobs\":0,\"metadata\":{},\"presence_penalty\":0.0,\"reasoning\":{\"effort\":\"none\"},\"created_at\":1788837162,\"store\":true,\"tools\":[],\"output\":[],\"top_p\":1.0,\"completed_at\":1788837162,\"previous_response_id\":\"resp_97f8cd28-c7ab-9d51-9c14-4fc531decdc8\",\"frequency_penalty\":0.0,\"parallel_tool_calls\":true,\"background\":false,\"temperature\":1.0,\"tool_choice\":\"auto\",\"model\":\"qwen3.8-max\",\"service_tier\":\"default\",\"id\":\"resp_1d81d34c-d3ec-9f98-ab59-5d4641d4217a\",\"max_output_tokens\":1024,\"object\":\"response\",\"status\":\"queued\"}}\n\nid:2\nevent:response.in_progress\n:HTTP_STATUS/200\ndata:{\"sequence_number\":1,\"type\":\"response.in_progress\",\"response\":{\"top_logprobs\":0,\"metadata\":{},\"presence_penalty\":0.0,\"reasoning\":{\"effort\":\"none\"},\"created_at\":1788837162,\"store\":true,\"tools\":[],\"output\":[],\"top_p\":1.0,\"completed_at\":1788837162,\"previous_response_id\":\"resp_97f8cd28-c7ab-9d51-9c14-4fc531decdc8\",\"frequency_penalty\":0.0,\"parallel_tool_calls\":true,\"background\":false,\"temperature\":1.0,\"tool_choice\":\"auto\",\"model\":\"qwen3.8-max\",\"service_tier\":\"default\",\"id\":\"resp_1d81d34c-d3ec-9f98-ab59-5d4641d4217a\",\"max_output_tokens\":1024,\"object\":\"response\",\"status\":\"in_progress\"}}\n\nid:3\nevent:response.output_item.added\n:HTTP_STATUS/200\ndata:{\"sequence_number\":2,\"item\":{\"id\":\"msg_7bc50c33-8fbd-4b10-91fc-4f2533bb2091\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[],\"status\":\"in_progress\"},\"output_index\":0,\"type\":\"response.output_item.added\"}\n\nid:4\nevent:response.content_part.added\n:HTTP_STATUS/200\ndata:{\"sequence_number\":3,\"output_index\":0,\"type\":\"response.content_part.added\",\"content_index\":0,\"item_id\":\"msg_7bc50c33-8fbd-4b10-91fc-4f2533bb2091\",\"part\":{\"type\":\"output_text\",\"annotations\":[],\"text\":\"\"}}\n\nid:5\nevent:response.output_text.delta\n:HTTP_STATUS/200\ndata:{\"sequence_number\":4,\"content_index\":0,\"item_id\":\"msg_7bc50c33-8fbd-4b10-91fc-4f2533bb2091\",\"delta\":\"ap\",\"output_index\":0,\"type\":\"response.output_text.delta\",\"logprobs\":[]}\n\nid:6\nevent:response.output_text.delta\n:HTTP_STATUS/200\ndata:{\"sequence_number\":5,\"content_index\":0,\"item_id\":\"msg_7bc50c33-8fbd-4b10-91fc-4f2533bb2091\",\"delta\":\"ricot\",\"output_index\":0,\"type\":\"response.output_text.delta\",\"logprobs\":[]}\n\nid:7\nevent:response.output_text.delta\n:HTTP_STATUS/200\ndata:{\"sequence_number\":6,\"content_index\":0,\"item_id\":\"msg_7bc50c33-8fbd-4b10-91fc-4f2533bb2091\",\"delta\":\"\",\"output_index\":0,\"type\":\"response.output_text.delta\",\"logprobs\":[]}\n\nid:8\nevent:response.output_text.done\n:HTTP_STATUS/200\ndata:{\"sequence_number\":7,\"content_index\":0,\"item_id\":\"msg_7bc50c33-8fbd-4b10-91fc-4f2533bb2091\",\"text\":\"apricot\",\"output_index\":0,\"type\":\"response.output_text.done\",\"logprobs\":[]}\n\nid:9\nevent:response.content_part.done\n:HTTP_STATUS/200\ndata:{\"sequence_number\":8,\"output_index\":0,\"type\":\"response.content_part.done\",\"content_index\":0,\"item_id\":\"msg_7bc50c33-8fbd-4b10-91fc-4f2533bb2091\",\"part\":{\"type\":\"output_text\",\"annotations\":[],\"text\":\"apricot\"}}\n\nid:10\nevent:response.output_item.done\n:HTTP_STATUS/200\ndata:{\"sequence_number\":9,\"item\":{\"id\":\"msg_7bc50c33-8fbd-4b10-91fc-4f2533bb2091\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[{\"type\":\"output_text\",\"annotations\":[],\"text\":\"apricot\"}],\"status\":\"completed\"},\"output_index\":0,\"type\":\"response.output_item.done\"}\n\nid:11\nevent:response.completed\n:HTTP_STATUS/200\ndata:{\"sequence_number\":10,\"type\":\"response.completed\",\"response\":{\"top_logprobs\":0,\"metadata\":{},\"presence_penalty\":0.0,\"reasoning\":{\"effort\":\"none\"},\"usage\":{\"total_tokens\":92,\"input_tokens_details\":{\"cached_tokens\":0},\"output_tokens\":3,\"input_tokens\":89,\"output_tokens_details\":{\"reasoning_tokens\":0},\"x_details\":[{\"total_tokens\":92,\"x_billing_type\":\"response_api\",\"output_tokens\":3,\"input_tokens\":89,\"prompt_tokens_details\":{\"cached_tokens\":0}}]},\"created_at\":1788837162,\"store\":true,\"tools\":[],\"output\":[{\"id\":\"msg_7bc50c33-8fbd-4b10-91fc-4f2533bb2091\",\"role\":\"assistant\",\"type\":\"message\",\"content\":[{\"type\":\"output_text\",\"annotations\":[],\"text\":\"apricot\"}],\"status\":\"completed\"}],\"top_p\":1.0,\"completed_at\":1788837162,\"previous_response_id\":\"resp_97f8cd28-c7ab-9d51-9c14-4fc531decdc8\",\"frequency_penalty\":0.0,\"parallel_tool_calls\":true,\"background\":false,\"temperature\":1.0,\"tool_choice\":\"auto\",\"model\":\"qwen3.8-max\",\"service_tier\":\"default\",\"id\":\"resp_1d81d34c-d3ec-9f98-ab59-5d4641d4217a\",\"max_output_tokens\":1024,\"object\":\"response\",\"status\":\"completed\"}}\n\n"
-      }
-    }
-  ]
-}

packages/ai/test/fixtures/recordings/openai-responses-websocket/reconstructs-full-context-after-reconnect.json

@@ -81,7 +81,7 @@
         {
           "direction": "client",
           "kind": "text",
-          "body": "{\"type\":\"response.create\",\"model\":\"gpt-5.5\",\"input\":[{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"Reply exactly: Alpha.\"}]},{\"type\":\"message\",\"id\":\"msg_ws_reconnect_1\",\"role\":\"assistant\",\"status\":\"completed\",\"content\":[{\"type\":\"output_text\",\"text\":\"Alpha.\"}]},{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"Reply exactly: Beta.\"}]}],\"store\":false,\"max_output_tokens\":30,\"include\":[\"reasoning.encrypted_content\"],\"reasoning\":{\"effort\":\"medium\",\"summary\":\"auto\"},\"text\":{\"verbosity\":\"low\"},\"instructions\":\"Follow the user's exact reply instruction.\"}"
+          "body": "{\"type\":\"response.create\",\"model\":\"gpt-5.5\",\"input\":[{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"Reply exactly: Alpha.\"}]},{\"type\":\"message\",\"id\":\"msg_ws_reconnect_1\",\"role\":\"assistant\",\"content\":[{\"type\":\"output_text\",\"text\":\"Alpha.\"}]},{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"Reply exactly: Beta.\"}]}],\"store\":false,\"max_output_tokens\":30,\"include\":[\"reasoning.encrypted_content\"],\"reasoning\":{\"effort\":\"medium\",\"summary\":\"auto\"},\"text\":{\"verbosity\":\"low\"},\"instructions\":\"Follow the user's exact reply instruction.\"}"
         },
         {
           "direction": "server",

packages/ai/test/fixtures/recordings/openai-responses-websocket/recovers-from-explicit-continuation-rejection.json

@@ -91,7 +91,7 @@
         {
           "direction": "client",
           "kind": "text",
-          "body": "{\"type\":\"response.create\",\"model\":\"gpt-5.5\",\"input\":[{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"Reply exactly: Ready.\"}]},{\"type\":\"message\",\"id\":\"msg_ws_rejection_1\",\"role\":\"assistant\",\"status\":\"completed\",\"content\":[{\"type\":\"output_text\",\"text\":\"Ready.\"}]},{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"Reply exactly: Recovered.\"}]}],\"store\":false,\"max_output_tokens\":30,\"include\":[\"reasoning.encrypted_content\"],\"reasoning\":{\"effort\":\"medium\",\"summary\":\"auto\"},\"text\":{\"verbosity\":\"low\"},\"instructions\":\"Follow the user's exact reply instruction.\"}"
+          "body": "{\"type\":\"response.create\",\"model\":\"gpt-5.5\",\"input\":[{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"Reply exactly: Ready.\"}]},{\"type\":\"message\",\"id\":\"msg_ws_rejection_1\",\"role\":\"assistant\",\"content\":[{\"type\":\"output_text\",\"text\":\"Ready.\"}]},{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"Reply exactly: Recovered.\"}]}],\"store\":false,\"max_output_tokens\":30,\"include\":[\"reasoning.encrypted_content\"],\"reasoning\":{\"effort\":\"medium\",\"summary\":\"auto\"},\"text\":{\"verbosity\":\"low\"},\"instructions\":\"Follow the user's exact reply instruction.\"}"
         },
         {
           "direction": "server",

packages/ai/test/provider-options/alibaba.types.ts

@@ -1,39 +0,0 @@
-import { LLM } from "../../src/index.js"
-import { Alibaba } from "../../src/providers.js"
-
-const provider = Alibaba.configure({ region: "ap-southeast-1" })
-Alibaba.configure({ baseURL: "https://gateway.example/v1" })
-Alibaba.configure({ region: "eu-central-1", workspaceID: "llm-workspace" })
-LLM.request({
-  model: provider.chat("qwen3.8-max"),
-  providerOptions: { reasoningEffort: "future", enableThinking: true, preserveThinking: false, toolStream: true },
-})
-LLM.request({
-  model: provider.messages("qwen3.8-max"),
-  providerOptions: { effort: "xhigh", thinking: { type: "enabled" } },
-})
-LLM.request({
-  model: provider.messages("qwen3.7-plus"),
-  providerOptions: { thinking: { type: "enabled", budgetTokens: 512 } },
-})
-LLM.request({
-  model: provider.responses("qwen3.8-max"),
-  providerOptions: { reasoningEffort: "low", enableThinking: true, store: true, previousResponseId: "resp_previous" },
-})
-// @ts-expect-error Region or complete base URL is required.
-Alibaba.configure({ apiKey: "fixture" })
-LLM.request({
-  model: provider.chat("qwen3.8-max"),
-  // @ts-expect-error Thinking toggle is a boolean.
-  providerOptions: { enableThinking: "true" },
-})
-LLM.request({
-  model: provider.messages("qwen3.8-max"),
-  // @ts-expect-error Messages uses effort.
-  providerOptions: { reasoningEffort: "high" },
-})
-LLM.request({
-  model: provider.responses("qwen3.8-max"),
-  // @ts-expect-error Responses does not use Chat thinking budgets.
-  providerOptions: { thinkingBudget: 512 },
-})

packages/ai/test/provider-package.test.ts

@@ -3,9 +3,6 @@ import { model } from "@opencode/ai/providers/openai"
 import { LLM } from "../src/index.js"
 import { Endpoint } from "../src/route/endpoint.js"
 
-const configuration = (provider: string, message: string) =>
-  expect.objectContaining({ _tag: "ProviderConfiguration", provider, message })
-
 describe("provider package entrypoints", () => {
   test("semantic API aliases expose the same contract", async () => {
     const modules = await Promise.all([
@@ -54,10 +51,6 @@ describe("provider package entrypoints", () => {
       import("@opencode/ai/providers/zai-coding-plan/chat"),
       import("@opencode/ai/providers/zai-coding-plan/messages"),
       import("@opencode/ai/providers/zai-coding-plan/responses"),
-      import("@opencode/ai/providers/alibaba"),
-      import("@opencode/ai/providers/alibaba/chat"),
-      import("@opencode/ai/providers/alibaba/messages"),
-      import("@opencode/ai/providers/alibaba/responses"),
     ])
 
     for (const module of modules) expect(module.model).toBeFunction()
@@ -68,34 +61,6 @@ describe("provider package entrypoints", () => {
     expect(modules[19].model).not.toBe(modules[20].model)
   })
 
-  test("maps Alibaba API entrypoints onto explicit regional routes", async () => {
-    const modules = await Promise.all([
-      import("@opencode/ai/providers/alibaba"),
-      import("@opencode/ai/providers/alibaba/chat"),
-      import("@opencode/ai/providers/alibaba/messages"),
-      import("@opencode/ai/providers/alibaba/responses"),
-    ])
-    expect(modules[0].model).toBe(modules[1].model)
-    const settings = {
-      region: "eu-central-1",
-      workspaceID: "llm-fixture",
-      apiKey: "fixture",
-      headers: { "x-test": "fixture" },
-      body: { extension: true },
-    }
-    const routes = ["alibaba-chat", "alibaba-chat", "alibaba-messages", "alibaba-responses"]
-    modules.forEach((module, index) => {
-      const model = module.model("qwen3.8-max", settings)
-      expect(model.provider).toBe("alibaba")
-      expect(model.route.id).toBe(routes[index])
-      expect(model.route.endpoint.baseURL).toBe(
-        `https://llm-fixture.eu-central-1.maas.aliyuncs.com/${index === 2 ? "apps/anthropic/v1" : "compatible-mode/v1"}`,
-      )
-      expect(model.route.defaults.headers).toEqual(settings.headers)
-      expect(model.route.defaults.http?.body).toEqual(settings.body)
-    })
-  })
-
   test("maps Moonshot API entrypoints onto provider-owned routes", async () => {
     const modules = await Promise.all([
       import("@opencode/ai/providers/moonshot"),
@@ -325,7 +290,7 @@ describe("provider package entrypoints", () => {
     const AnthropicCompatible = await import("@opencode/ai/providers/anthropic-compatible")
     expect(() =>
       Reflect.apply(AnthropicCompatible.model, undefined, ["compatible-model", { apiKey: "fixture" }]),
-    ).toThrow(configuration("anthropic-compatible", "Anthropic-compatible providers require a baseURL"))
+    ).toThrow("Anthropic-compatible providers require a baseURL")
   })
 
   test("rejects conflicting Anthropic-compatible auth settings at runtime", async () => {
@@ -340,10 +305,10 @@ describe("provider package entrypoints", () => {
           baseURL: "https://messages.example.test/v1",
         },
       ]),
-    ).toThrow(configuration("anthropic-compatible", "Anthropic-compatible apiKey cannot be combined with authToken"))
+    ).toThrow("Anthropic-compatible apiKey cannot be combined with authToken")
     expect(() =>
       Reflect.apply(Anthropic.model, undefined, ["claude-sonnet-4-6", { apiKey: "fixture", authToken: "token" }]),
-    ).toThrow(configuration("anthropic", "Anthropic apiKey cannot be combined with authToken"))
+    ).toThrow("Anthropic apiKey cannot be combined with authToken")
   })
 
   test("maps legacy OpenAI organization and project settings to headers", () => {
@@ -493,45 +458,43 @@ describe("provider package entrypoints", () => {
         "gemini-3.5-flash",
         { accessToken: "token", apiKey: "fixture", project: "vertex-project" },
       ]),
-    ).toThrow(configuration("google-vertex", "Google Vertex apiKey cannot be combined with accessToken or auth"))
+    ).toThrow("Google Vertex apiKey cannot be combined with accessToken or auth")
     const configured = Reflect.apply(GoogleVertex.configure, undefined, [
       { accessToken: "token", auth: {}, project: "vertex-project" },
     ])
-    expect(() => configured.model("gemini-3.5-flash")).toThrow(
-      configuration("google-vertex", "Google Vertex accessToken cannot be combined with auth"),
-    )
+    expect(() => configured.model("gemini-3.5-flash")).toThrow("Google Vertex accessToken cannot be combined with auth")
     expect(() =>
       Reflect.apply(GoogleVertexMessages.model, undefined, [
         "claude-sonnet-4-6",
         { apiKey: "fixture", project: "vertex-project" },
       ]),
-    ).toThrow(configuration("google-vertex", "Google Vertex Messages does not support API keys"))
+    ).toThrow("Google Vertex Messages does not support API keys")
     expect(() =>
       Reflect.apply(Providers.GoogleVertexMessages.configure, undefined, [
         { apiKey: "fixture", project: "vertex-project" },
       ]),
-    ).toThrow(configuration("google-vertex", "Google Vertex Messages does not support API keys"))
+    ).toThrow("Google Vertex Messages does not support API keys")
     expect(() =>
       Reflect.apply(GoogleVertexChat.model, undefined, [
         "deepseek-ai/deepseek-v3.2-maas",
         { apiKey: "fixture", project: "vertex-project" },
       ]),
-    ).toThrow(configuration("google-vertex", "Google Vertex Chat does not support API keys"))
+    ).toThrow("Google Vertex Chat does not support API keys")
     expect(() =>
       Reflect.apply(Providers.GoogleVertexChat.configure, undefined, [
         { apiKey: "fixture", project: "vertex-project" },
       ]),
-    ).toThrow(configuration("google-vertex", "Google Vertex Chat does not support API keys"))
+    ).toThrow("Google Vertex Chat does not support API keys")
     expect(() =>
       Reflect.apply(GoogleVertexResponses.model, undefined, [
         "xai/grok-4.20-reasoning",
         { apiKey: "fixture", project: "vertex-project" },
       ]),
-    ).toThrow(configuration("google-vertex", "Google Vertex Responses does not support API keys"))
+    ).toThrow("Google Vertex Responses does not support API keys")
     expect(() =>
       Reflect.apply(Providers.GoogleVertexResponses.configure, undefined, [
         { apiKey: "fixture", project: "vertex-project" },
       ]),
-    ).toThrow(configuration("google-vertex", "Google Vertex Responses does not support API keys"))
+    ).toThrow("Google Vertex Responses does not support API keys")
   })
 })

packages/ai/test/provider/alibaba-capabilities.recorded.test.ts

@@ -1,254 +0,0 @@
-import { describe, expect } from "bun:test"
-import { Effect } from "effect"
-import { LLM, LLMEvent, LLMRequest, Message, ToolDefinition } from "../../src/index.js"
-import { Alibaba } from "../../src/providers.js"
-import { LLMClient } from "../../src/route.js"
-import { compileRequest } from "../../src/route/client.js"
-import { recordedTests } from "../recorded-test.js"
-
-const alibaba = Alibaba.configure({ region: "ap-southeast-1", apiKey: process.env.ALIBABA_API_KEY ?? "fixture" })
-const record = (api: "chat" | "messages" | "responses") =>
-  recordedTests({
-    prefix: `alibaba-${api}`,
-    provider: "alibaba",
-    protocol: `alibaba-${api}`,
-    requires: ["ALIBABA_API_KEY"],
-    tags: ["region:ap-southeast-1"],
-  })
-
-for (const api of ["chat", "messages", "responses"] as const) {
-  const recorded = record(api)
-  describe(`Alibaba ${api} capabilities`, () => {
-    for (const enabled of [false, true]) {
-      recorded.effect.with(
-        `Qwen 3.7 Plus streams thinking ${enabled ? "enabled" : "disabled"}`,
-        { tags: ["thinking", "usage"] },
-        () =>
-          Effect.gen(function* () {
-            const request = LLM.request({
-              model: alibaba[api]("qwen3.7-plus"),
-              providerOptions:
-                api === "messages"
-                  ? { thinking: { type: enabled ? "enabled" : "disabled", ...(enabled ? { budgetTokens: 1024 } : {}) } }
-                  : api === "chat"
-                    ? { enableThinking: enabled, ...(enabled ? { thinkingBudget: 1024 } : {}) }
-                    : { enableThinking: enabled },
-              prompt: "What is 173 multiplied by 219? Reply with only the final integer.",
-              generation: { maxTokens: 4096 },
-            })
-            const compiled = yield* compileRequest(request)
-            expect(api === "messages" ? compiled.body.thinking.type : compiled.body.enable_thinking).toBe(
-              api === "messages" ? (enabled ? "enabled" : "disabled") : enabled,
-            )
-            const response = yield* LLMClient.generate(request)
-            expect(response.text.replaceAll(",", "")).toContain("37887")
-            expect(response.reasoning.length > 0).toBe(enabled)
-            expect(response.finishReason.normalized).toBe("stop")
-            expect(response.usage?.inputTokens).toBeGreaterThan(0)
-            expect(response.usage?.outputTokens).toBeGreaterThan(0)
-          }),
-        120_000,
-      )
-    }
-    recorded.effect.with(
-      "Qwen 3.8 Flash reads image bytes",
-      { tags: ["image"] },
-      () =>
-        Effect.gen(function* () {
-          const bytes = yield* Effect.promise(() =>
-            Bun.file(new URL("../fixtures/media/restroom.png", import.meta.url)).bytes(),
-          )
-          const response = yield* LLMClient.generate(
-            LLM.request({
-              model: alibaba[api]("qwen3.8-flash"),
-              providerOptions: api === "messages" ? { thinking: { type: "disabled" } } : { enableThinking: false },
-              messages: [
-                Message.user([
-                  { type: "text", text: "Read the three words in this image. Reply with only the words in order." },
-                  { type: "media", mediaType: "image/png", data: bytes },
-                ]),
-              ],
-              generation: { maxTokens: 4096 },
-            }),
-          )
-          expect(response.text.toLowerCase()).toContain("jiggling restroom prison")
-          expect(response.finishReason.normalized).toBe("stop")
-        }),
-      120_000,
-    )
-    recorded.effect.with(
-      "Qwen 3.8 Max obeys named tool choice",
-      { tags: ["tool", "tool-choice"] },
-      () =>
-        Effect.gen(function* () {
-          const response = yield* LLMClient.generate(
-            LLM.request({
-              model: alibaba[api]("qwen3.8-max"),
-              prompt: "Find the current weather in Paris.",
-              providerOptions: api === "messages" ? { thinking: { type: "disabled" } } : { reasoningEffort: "none" },
-              tools: [
-                ToolDefinition.make({
-                  name: "get_weather",
-                  description: "Get weather in a city",
-                  inputSchema: {
-                    type: "object",
-                    properties: { city: { type: "string", enum: ["Paris"] } },
-                    required: ["city"],
-                  },
-                }),
-              ],
-              toolChoice: { type: "tool", name: "get_weather" },
-              generation: { maxTokens: 4096 },
-            }),
-          )
-          expect(response.toolCalls).toMatchObject([{ name: "get_weather", input: { city: "Paris" } }])
-          expect(response.finishReason.normalized).toBe(api === "messages" ? "stop" : "tool-calls")
-          if (api === "messages") expect(response.finishReason.raw).toBe("end_turn")
-        }),
-      120_000,
-    )
-  })
-}
-
-record("chat").effect.with(
-  "Qwen 3.8 Max returns a JSON object",
-  { tags: ["structured-output"] },
-  () =>
-    Effect.gen(function* () {
-      const response = yield* LLMClient.generate(
-        LLM.request({
-          model: alibaba.chat("qwen3.8-max"),
-          prompt: 'Return a JSON object with one key "city" set to the capital city of France.',
-          providerOptions: { reasoningEffort: "none", responseFormat: { type: "json_object" } },
-          generation: { maxTokens: 1024 },
-        }),
-      )
-      expect(JSON.parse(response.text)).toEqual({ city: "Paris" })
-      expect(response.finishReason.normalized).toBe("stop")
-    }),
-  120_000,
-)
-
-record("messages").effect.with(
-  "Qwen 3.8 Max follows a JSON schema",
-  { tags: ["structured-output"] },
-  () =>
-    Effect.gen(function* () {
-      const response = yield* LLMClient.generate(
-        LLM.request({
-          model: alibaba.messages("qwen3.8-max"),
-          prompt: 'Return a JSON object with one key "city" set to the capital city of France.',
-          providerOptions: {
-            thinking: { type: "disabled" },
-            outputConfig: {
-              format: {
-                type: "json_schema",
-                schema: {
-                  type: "object",
-                  properties: { city: { type: "string" } },
-                  required: ["city"],
-                  additionalProperties: false,
-                },
-              },
-            },
-          },
-          generation: { maxTokens: 1024 },
-        }),
-      )
-      expect(JSON.parse(response.text)).toEqual({ city: "Paris" })
-      expect(response.finishReason.normalized).toBe("stop")
-    }),
-  120_000,
-)
-
-const responses = record("responses")
-responses.effect.with(
-  "Qwen 3.8 Max continues a stored response",
-  { tags: ["continuation", "storage"] },
-  () =>
-    Effect.gen(function* () {
-      const first = yield* LLMClient.generate(
-        LLM.request({
-          model: alibaba.responses("qwen3.8-max"),
-          prompt: "Remember the password word apricot. Reply OK.",
-          providerOptions: { store: true, reasoningEffort: "none" },
-          generation: { maxTokens: 1024 },
-        }),
-      )
-      const id = first.events.find(LLMEvent.is.finish)?.providerMetadata?.alibaba?.responseId
-      expect(id).toBeString()
-      if (typeof id !== "string") throw new Error("Missing Alibaba response ID")
-      const second = yield* LLMClient.generate(
-        LLM.request({
-          model: alibaba.responses("qwen3.8-max"),
-          prompt: "What word did I ask you to remember? Reply with only the word.",
-          providerOptions: { previousResponseId: id, store: true, reasoningEffort: "none" },
-          generation: { maxTokens: 1024 },
-        }),
-      )
-      expect(second.text.toLowerCase()).toContain("apricot")
-      expect(second.finishReason.normalized).toBe("stop")
-    }),
-  120_000,
-)
-
-responses.effect.with(
-  "Qwen 3.8 Max uses hosted web search and extraction",
-  { tags: ["hosted-tool", "web-search", "web-extractor"] },
-  () =>
-    Effect.gen(function* () {
-      const request = LLM.request({
-        model: alibaba.responses("qwen3.8-max"),
-        prompt:
-          "Use web search to find Alibaba Cloud Model Studio's official documentation, then use web_extractor to read the page. Give a brief summary with the source URL.",
-        tools: [Alibaba.webSearch(), Alibaba.webExtractor()],
-        providerOptions: { reasoningEffort: "low", store: false },
-        generation: { maxTokens: 4096 },
-      })
-      const response = yield* LLMClient.generate(request)
-      expect(response.toolCalls).toEqual(
-        expect.arrayContaining([
-          expect.objectContaining({ name: "web_search", providerExecuted: true }),
-          expect.objectContaining({ name: "web_extractor", providerExecuted: true }),
-        ]),
-      )
-      expect(response.text.toLowerCase()).toContain("alibaba")
-      expect(response.events.some(LLMEvent.is.toolResult)).toBe(true)
-      const replay = yield* compileRequest(
-        LLMRequest.update(request, {
-          messages: [...request.messages, response.message, Message.user("Summarize in one sentence.")],
-        }),
-      )
-      expect(replay.body.input).toEqual(
-        expect.arrayContaining([
-          expect.objectContaining({ type: "web_search_call" }),
-          expect.objectContaining({ type: "web_extractor_call" }),
-        ]),
-      )
-    }),
-  180_000,
-)
-
-responses.effect.with(
-  "Qwen 3.8 Max uses hosted code interpreter",
-  { tags: ["hosted-tool", "code-interpreter"] },
-  () =>
-    Effect.gen(function* () {
-      const response = yield* LLMClient.generate(
-        LLM.request({
-          model: alibaba.responses("qwen3.8-max"),
-          prompt:
-            "Use the code interpreter to compute the SHA-256 hash of the UTF-8 string hello (no newline). Reply with only the hash.",
-          tools: [Alibaba.codeInterpreter()],
-          providerOptions: { reasoningEffort: "low" },
-          generation: { maxTokens: 4096 },
-        }),
-      )
-      expect(response.toolCalls).toEqual(
-        expect.arrayContaining([expect.objectContaining({ name: "code_interpreter", providerExecuted: true })]),
-      )
-      expect(response.text).toContain("2cf24dba5fb0a30e26e83b2ac5b9e29e1b161e5c1fa7425e73043362938b9824")
-      expect(response.events.some(LLMEvent.is.toolResult)).toBe(true)
-    }),
-  180_000,
-)

packages/ai/test/provider/alibaba.recorded.test.ts

@@ -1,163 +0,0 @@
-import { describe, expect } from "bun:test"
-import { Effect } from "effect"
-import { LLM, LLMEvent, LLMRequest, Message, ToolDefinition, type LLMResponse } from "../../src/index.js"
-import { Alibaba } from "../../src/providers.js"
-import { LLMClient } from "../../src/route.js"
-import { compileRequest } from "../../src/route/client.js"
-import { recordedTests } from "../recorded-test.js"
-
-const alibaba = Alibaba.configure({ region: "ap-southeast-1", apiKey: process.env.ALIBABA_API_KEY ?? "fixture" })
-const weather = ToolDefinition.make({
-  name: "get_weather",
-  description: "Get the current weather in a city",
-  inputSchema: {
-    type: "object",
-    properties: { city: { type: "string", enum: ["Paris"] } },
-    required: ["city"],
-    additionalProperties: false,
-  },
-})
-
-for (const api of ["chat", "messages", "responses"] as const) {
-  const recorded = recordedTests({
-    prefix: `alibaba-${api}`,
-    provider: "alibaba",
-    protocol: `alibaba-${api}`,
-    requires: ["ALIBABA_API_KEY"],
-    tags: ["region:ap-southeast-1"],
-  })
-  describe(`Alibaba ${api}`, () => {
-    for (const effort of api === "messages"
-      ? [undefined, "low", "medium", "high", "xhigh", "max"]
-      : [undefined, "none", "minimal", "low", "medium", "high", "xhigh", "max"]) {
-      recorded.effect.with(
-        `Qwen 3.8 Max streams ${effort ?? "default"} effort`,
-        { tags: ["text", "reasoning", "usage"] },
-        () =>
-          Effect.gen(function* () {
-            const request = LLM.request({
-              model: alibaba[api]("qwen3.8-max"),
-              prompt: "What is 173 multiplied by 219? Reply with only the final integer.",
-              providerOptions: api === "messages" ? { effort } : { reasoningEffort: effort },
-              generation: { maxTokens: 4096 },
-            })
-            const compiled = yield* compileRequest(request)
-            expect(compiled.body.enable_thinking).toBeUndefined()
-            expect(compiled.body.thinking).toBeUndefined()
-            expect(
-              api === "chat"
-                ? compiled.body.reasoning_effort
-                : api === "messages"
-                  ? compiled.body.output_config?.effort
-                  : compiled.body.reasoning?.effort,
-            ).toBe(effort)
-            const response = yield* LLMClient.generate(request)
-            expect(response.text.replaceAll(",", "")).toContain("37887")
-            expect(response.reasoning.length > 0).toBe(effort !== "none")
-            expect(response.finishReason.normalized).toBe("stop")
-            expectUsage(response)
-          }),
-        120_000,
-      )
-    }
-
-    recorded.effect.with(
-      "Qwen 3.8 Max replays reasoning through a tool loop and follow-up",
-      { tags: ["tool", "tool-loop", "reasoning", "continuation"] },
-      () =>
-        Effect.gen(function* () {
-          const request = LLM.request({
-            model: alibaba[api]("qwen3.8-max"),
-            providerOptions:
-              api === "messages"
-                ? { effort: "medium" }
-                : api === "chat"
-                  ? { reasoningEffort: "medium", preserveThinking: true, toolStream: true }
-                  : { reasoningEffort: "medium", store: false },
-            prompt:
-              "We have a budget of 38000 dollars for 219 trips costing 173 dollars each. Calculate whether that is affordable. If it is, use get_weather to look up the current weather in Paris. After receiving the result, report the weather in one short sentence.",
-            tools: [weather],
-            generation: { maxTokens: 4096 },
-          })
-          const first = yield* LLMClient.generate(request)
-          expect(first.toolCalls).toMatchObject([{ name: "get_weather", input: { city: "Paris" } }])
-          expect(first.finishReason.normalized).toBe("tool-calls")
-          expect(first.reasoning.length).toBeGreaterThan(0)
-          expect(first.events.some(LLMEvent.is.toolInputDelta)).toBe(true)
-          expectUsage(first)
-          const continuation = LLMRequest.update(request, {
-            messages: [
-              ...request.messages,
-              first.message,
-              ...first.toolCalls.map((call) =>
-                Message.tool({ id: call.id, name: call.name, result: { condition: "sunny", temperature: "18C" } }),
-              ),
-            ],
-          })
-          expectReasoning(api, (yield* compileRequest(continuation)).body, first)
-          const second = yield* LLMClient.generate(continuation)
-          expect(second.text.toLowerCase()).toContain("sunny")
-          expect(second.toolCalls).toHaveLength(0)
-          expect(second.finishReason.normalized).toBe("stop")
-          expectUsage(second)
-          const followUp = LLMRequest.update(continuation, {
-            messages: [
-              ...continuation.messages,
-              second.message,
-              Message.user("What temperature did the tool report? Reply with only the temperature."),
-            ],
-          })
-          expectReasoning(api, (yield* compileRequest(followUp)).body, first)
-          const third = yield* LLMClient.generate(followUp)
-          expect(third.text).toContain("18")
-          expect(third.toolCalls).toHaveLength(0)
-          expect(third.finishReason.normalized).toBe("stop")
-          expectUsage(third)
-        }),
-      180_000,
-    )
-  })
-}
-
-function expectUsage(response: LLMResponse) {
-  expect(response.usage?.inputTokens).toBeGreaterThan(0)
-  expect(response.usage?.outputTokens).toBeGreaterThan(0)
-  expect(response.events.filter(LLMEvent.is.finish)).toHaveLength(1)
-}
-
-function expectReasoning(
-  api: "chat" | "messages" | "responses",
-  body: Readonly<Record<string, unknown>>,
-  response: LLMResponse,
-) {
-  if (api === "chat") {
-    expect(body.messages).toEqual(
-      expect.arrayContaining([expect.objectContaining({ role: "assistant", reasoning_content: response.reasoning })]),
-    )
-    return
-  }
-  for (const part of response.message.content.filter((part) => part.type === "reasoning")) {
-    if (api === "messages") {
-      expect(body.messages).toEqual(
-        expect.arrayContaining([
-          expect.objectContaining({
-            role: "assistant",
-            content: expect.arrayContaining([
-              { type: "thinking", thinking: part.text, signature: part.providerMetadata?.alibaba?.signature ?? "" },
-            ]),
-          }),
-        ]),
-      )
-      continue
-    }
-    expect(body.input).toEqual(
-      expect.arrayContaining([
-        expect.objectContaining({
-          type: "reasoning",
-          id: part.providerMetadata?.alibaba?.itemId,
-          summary: expect.arrayContaining([{ type: "summary_text", text: part.text }]),
-        }),
-      ]),
-    )
-  }
-}

packages/ai/test/provider/alibaba.test.ts

@@ -1,382 +0,0 @@
-import { expect, test } from "bun:test"
-import { ConfigProvider, Effect } from "effect"
-import { Headers } from "effect/unstable/http"
-import { Auth, LLM, LLMClient, LLMRequest, Message, ReasoningPart, ToolDefinition } from "../../src/index.js"
-import { Alibaba } from "../../src/providers.js"
-import { compileRequest } from "../../src/route/client.js"
-import { Endpoint } from "../../src/route/endpoint.js"
-import { it } from "../lib/effect.js"
-import { dynamicResponse } from "../lib/http.js"
-import { sseEvents } from "../lib/sse.js"
-
-const tool = ToolDefinition.make({
-  name: "lookup",
-  description: "Look up a value",
-  inputSchema: { type: "object", properties: { key: { type: "string" } } },
-})
-const paths = {
-  chat: "/compatible-mode/v1/chat/completions",
-  messages: "/apps/anthropic/v1/messages",
-  responses: "/compatible-mode/v1/responses",
-}
-
-it.effect("Alibaba owns regional shared and workspace-specific endpoints", () =>
-  Effect.gen(function* () {
-    for (const region of [
-      "ap-southeast-1",
-      "cn-beijing",
-      "cn-hongkong",
-      "us-east-1",
-      "eu-central-1",
-      "ap-northeast-1",
-    ]) {
-      const provider = Alibaba.configure({ region, workspaceID: "llm-fixture", apiKey: "fixture" })
-      expect(provider.model).toBe(provider.chat)
-      for (const api of ["chat", "messages", "responses"] as const) {
-        const model = provider[api]("qwen-plus-us")
-        const request = LLM.request({ model, prompt: "Hello" })
-        const compiled = yield* compileRequest(request)
-        expect(Endpoint.render(model.route.endpoint, { request, body: compiled.body }).toString()).toBe(
-          `https://llm-fixture.${region}.maas.aliyuncs.com${paths[api]}`,
-        )
-        expect(model.provider).toBe("alibaba")
-        expect(model.route.id).toBe(`alibaba-${api}`)
-        expect(compiled.body.model).toBe("qwen-plus-us")
-        for (const field of [
-          "thinking",
-          "enable_thinking",
-          "thinking_budget",
-          "preserve_thinking",
-          "reasoning_effort",
-          "reasoning",
-          "output_config",
-          "store",
-          "tool_stream",
-        ])
-          expect(compiled.body[field]).toBeUndefined()
-      }
-    }
-    for (const [region, host] of [
-      ["ap-southeast-1", "dashscope-intl.aliyuncs.com"],
-      ["cn-beijing", "dashscope.aliyuncs.com"],
-      ["cn-hongkong", "cn-hongkong.dashscope.aliyuncs.com"],
-      ["us-east-1", "dashscope-us.aliyuncs.com"],
-    ]) {
-      const provider = Alibaba.configure({ region, apiKey: "fixture" })
-      for (const api of ["chat", "messages", "responses"] as const) {
-        const request = LLM.request({ model: provider[api]("qwen3.8-max") })
-        const compiled = yield* compileRequest(request)
-        expect(Endpoint.render(request.model.route.endpoint, { request, body: compiled.body }).toString()).toBe(
-          `https://${host}${paths[api]}`,
-        )
-      }
-    }
-  }),
-)
-
-test("Alibaba requires explicit placement and supports complete base URL overrides", () => {
-  for (const region of ["eu-central-1", "ap-northeast-1", "future-region"])
-    expect(() => Alibaba.configure({ region })).toThrow(
-      expect.objectContaining({
-        _tag: "ProviderConfiguration",
-        provider: "alibaba",
-        message: `Alibaba region ${region} requires workspaceID or baseURL`,
-      }),
-    )
-  for (const config of [
-    { baseURL: "https://gateway.example/prefix" },
-    { region: "future-region", workspaceID: "ignored", baseURL: "https://gateway.example/prefix" },
-  ]) {
-    const provider = Alibaba.configure(config)
-    for (const api of ["chat", "messages", "responses"] as const)
-      expect(provider[api]("unchanged-id").route.endpoint.baseURL).toBe(config.baseURL)
-  }
-  expect(
-    Alibaba.configure({ region: "future-region", workspaceID: "llm-fixture" }).chat("new-model").route.endpoint.baseURL,
-  ).toBe("https://llm-fixture.future-region.maas.aliyuncs.com/compatible-mode/v1")
-})
-
-it.effect("Alibaba resolves explicit auth, API keys, and regional environment credentials", () =>
-  Effect.gen(function* () {
-    for (const item of [
-      {
-        config: {},
-        env: { DASHSCOPE_API_KEY: "primary", ALIBABA_API_KEY: "fallback" },
-        headers: { authorization: "Bearer primary" },
-      },
-      { config: {}, env: { ALIBABA_API_KEY: "fallback" }, headers: { authorization: "Bearer fallback" } },
-      {
-        config: { apiKey: "explicit" },
-        env: { DASHSCOPE_API_KEY: "primary" },
-        headers: { authorization: "Bearer explicit" },
-      },
-      {
-        config: { auth: Auth.header("x-custom-key", "custom") },
-        env: { DASHSCOPE_API_KEY: "primary" },
-        headers: { "x-custom-key": "custom" },
-      },
-    ]) {
-      const provider = Alibaba.configure({ region: "ap-southeast-1", ...item.config })
-      for (const api of ["chat", "messages", "responses"] as const) {
-        const request = LLM.request({ model: provider[api]("qwen3.8-max") })
-        const headers = yield* request.model.route.auth
-          .apply({ request, method: "POST", url: "https://fixture", body: "{}", headers: Headers.empty })
-          .pipe(Effect.provide(ConfigProvider.layer(ConfigProvider.fromEnv({ env: item.env }))))
-        expect(headers).toEqual(expect.objectContaining(item.headers))
-      }
-    }
-  }),
-)
-
-it.effect("Alibaba keeps native reasoning controls and future efforts on their selected API", () =>
-  Effect.gen(function* () {
-    const provider = Alibaba.configure({ region: "ap-southeast-1", apiKey: "fixture" })
-    for (const effort of ["none", "minimal", "low", "medium", "high", "xhigh", "max", "future-effort"]) {
-      const chat = yield* compileRequest(
-        LLM.request({ model: provider.chat("qwen3.8-max"), providerOptions: { reasoningEffort: effort } }),
-      )
-      const messages = yield* compileRequest(
-        LLM.request({ model: provider.messages("qwen3.8-max"), providerOptions: { effort } }),
-      )
-      const responses = yield* compileRequest(
-        LLM.request({ model: provider.responses("qwen3.8-max"), providerOptions: { reasoningEffort: effort } }),
-      )
-      expect(chat.body.reasoning_effort).toBe(effort)
-      expect(messages.body.output_config).toEqual({ effort })
-      expect(responses.body.reasoning).toEqual({ effort })
-      for (const result of [chat, messages, responses]) {
-        expect(result.body.thinking).toBeUndefined()
-        expect(result.body.enable_thinking).toBeUndefined()
-      }
-    }
-    for (const enableThinking of [true, false]) {
-      const chat = yield* compileRequest(
-        LLM.request({
-          model: provider.chat("qwen3.7-plus"),
-          tools: [tool],
-          generation: { maxTokens: 1234, topK: 20 },
-          providerOptions: {
-            enableThinking,
-            thinkingBudget: 512,
-            preserveThinking: false,
-            clearThinking: false,
-            toolStream: false,
-            parallelToolCalls: false,
-            repetitionPenalty: 1.1,
-            responseFormat: { type: "json_object" },
-            enableSearch: true,
-            searchOptions: { forced_search: true, search_strategy: "future-strategy", enable_search_extension: false },
-          },
-        }),
-      )
-      expect(chat.body).toMatchObject({
-        enable_thinking: enableThinking,
-        thinking_budget: 512,
-        preserve_thinking: false,
-        clear_thinking: false,
-        tool_stream: false,
-        parallel_tool_calls: false,
-        repetition_penalty: 1.1,
-        max_completion_tokens: 1234,
-        top_k: 20,
-        response_format: { type: "json_object" },
-        enable_search: true,
-        search_options: { forced_search: true, search_strategy: "future-strategy", enable_search_extension: false },
-      })
-      expect(chat.body.max_tokens).toBeUndefined()
-      expect(chat.body.tools).toEqual([
-        expect.objectContaining({ function: expect.not.objectContaining({ strict: expect.anything() }) }),
-      ])
-      const responses = yield* compileRequest(
-        LLM.request({
-          model: provider.responses("qwen3.7-plus"),
-          providerOptions: { enableThinking, store: false, previousResponseId: "resp_previous" },
-        }),
-      )
-      expect(responses.body).toMatchObject({
-        enable_thinking: enableThinking,
-        store: false,
-        previous_response_id: "resp_previous",
-      })
-    }
-    for (const thinking of [
-      { type: "enabled" },
-      { type: "disabled" },
-      { type: "enabled", budgetTokens: 512 },
-      { type: "future", budget_tokens: 4096 },
-    ]) {
-      const messages = yield* compileRequest(
-        LLM.request({ model: provider.messages("qwen3.7-plus"), providerOptions: { thinking } }),
-      )
-      expect(messages.body.thinking).toEqual({
-        type: thinking.type,
-        budget_tokens: thinking.budgetTokens ?? thinking.budget_tokens,
-      })
-    }
-  }),
-)
-
-it.effect("Alibaba validates malformed options before execution", () =>
-  Effect.gen(function* () {
-    const provider = Alibaba.configure({ region: "ap-southeast-1", apiKey: "fixture" })
-    for (const [api, providerOptions] of [
-      ["chat", { preserveThinking: "false" }],
-      ["messages", { thinking: { type: "enabled", budgetTokens: "512" } }],
-      ["responses", { enableThinking: "false" }],
-    ] as const) {
-      const model = provider[api]("qwen3.8-max").route.with({ providerOptions }).model({ id: "qwen3.8-max" })
-      const error = yield* compileRequest(LLM.request({ model })).pipe(Effect.flip)
-      expect(error.reason._tag).toBe("InvalidRequest")
-    }
-  }),
-)
-
-it.effect("Alibaba preserves unsigned and signed Messages thinking without a budget requirement", () =>
-  Effect.gen(function* () {
-    const result = yield* compileRequest(
-      LLM.request({
-        model: Alibaba.configure({ region: "ap-southeast-1", apiKey: "fixture" }).messages("qwen3.8-max"),
-        providerOptions: { thinking: { type: "enabled" }, effort: "low" },
-        messages: [
-          Message.user("Hello"),
-          Message.assistant([
-            ReasoningPart.make({ type: "reasoning", text: "unsigned" }),
-            ReasoningPart.make({
-              type: "reasoning",
-              text: "signed",
-              providerMetadata: { alibaba: { signature: "opaque" } },
-            }),
-          ]),
-        ],
-      }),
-    )
-    expect(result.body.thinking).toEqual({ type: "enabled" })
-    expect(result.body.messages).toEqual(
-      expect.arrayContaining([
-        expect.objectContaining({
-          role: "assistant",
-          content: [
-            { type: "thinking", thinking: "unsigned", signature: "" },
-            { type: "thinking", thinking: "signed", signature: "opaque" },
-          ],
-        }),
-      ]),
-    )
-  }),
-)
-
-it.effect("Alibaba serializes Messages configuration, per-request controls, and final HTTP overlays", () =>
-  LLMClient.generate(
-    LLM.request({
-      model: Alibaba.configure({
-        baseURL: "https://gateway.example/v1",
-        apiKey: "fixture",
-        providerOptions: { thinking: { type: "enabled", budgetTokens: 512 }, effort: "high" },
-      }).messages("qwen3.8-max"),
-      prompt: "Hello",
-      providerOptions: { effort: "low" },
-      http: { body: { output_config: { effort: "medium" }, extension: true } },
-    }),
-  ).pipe(
-    Effect.provide(
-      dynamicResponse((input) =>
-        Effect.sync(() => {
-          expect(input.request.url).toBe("https://gateway.example/v1/messages")
-          expect(input.request.headers.authorization).toBe("Bearer fixture")
-          expect(input.request.headers["anthropic-version"]).toBe("2023-06-01")
-          expect(JSON.parse(input.text)).toMatchObject({
-            thinking: { type: "enabled", budget_tokens: 512 },
-            output_config: { effort: "medium" },
-            extension: true,
-          })
-          return input.respond(
-            sseEvents(
-              {
-                type: "message_start",
-                message: { id: "msg_fixture", content: [], usage: { input_tokens: 1, output_tokens: 0 } },
-              },
-              { type: "message_delta", delta: { stop_reason: "end_turn" }, usage: { output_tokens: 1 } },
-              { type: "message_stop" },
-            ),
-            { headers: { "content-type": "text/event-stream" } },
-          )
-        }),
-      ),
-    ),
-  ),
-)
-
-it.effect("Alibaba Responses lowers hosted tools and named selection using HTTP even with a WebSocket executor", () =>
-  Effect.gen(function* () {
-    const request = LLM.request({
-      model: Alibaba.configure({ region: "ap-southeast-1", apiKey: "fixture" }).responses("qwen3.8-max"),
-      tools: [Alibaba.webSearch(), Alibaba.webExtractor(), Alibaba.codeInterpreter(), tool],
-      prompt: "Hello",
-    })
-    const compiled = yield* compileRequest(request)
-    expect(compiled.body.tools).toMatchObject([
-      { type: "web_search" },
-      { type: "web_extractor" },
-      { type: "code_interpreter" },
-      { type: "function", name: "lookup" },
-    ])
-    const named = yield* compileRequest(
-      LLMRequest.update(request, { tools: [tool], toolChoice: { type: "tool", name: "lookup" } }),
-    )
-    expect(named.body.tool_choice).toEqual({
-      type: "allowed_tools",
-      mode: "required",
-      tools: [{ type: "function", name: "lookup" }],
-    })
-    const response = yield* LLMClient.generate(request, {
-      webSocket: { execute: () => Effect.die("Unexpected WebSocket") },
-    })
-    expect(response.toolCalls).toMatchObject([
-      {
-        name: "web_extractor",
-        providerExecuted: true,
-        input: { urls: ["https://example.com"], goal: "Read the page" },
-      },
-    ])
-    const replay = yield* compileRequest(
-      LLMRequest.update(request, { messages: [...request.messages, response.message, Message.user("Continue")] }),
-    )
-    expect(replay.body.input).toEqual(
-      expect.arrayContaining([
-        expect.objectContaining({
-          type: "web_extractor_call",
-          id: "extract_1",
-          urls: ["https://example.com"],
-          goal: "Read the page",
-          result: { text: "fixture" },
-        }),
-      ]),
-    )
-  }).pipe(
-    Effect.provide(
-      dynamicResponse((input) =>
-        Effect.sync(() => {
-          expect(input.request.url).toBe("https://dashscope-intl.aliyuncs.com/compatible-mode/v1/responses")
-          return input.respond(
-            sseEvents(
-              {
-                type: "response.output_item.done",
-                output_index: 0,
-                item: {
-                  type: "web_extractor_call",
-                  id: "extract_1",
-                  status: "completed",
-                  urls: ["https://example.com"],
-                  goal: "Read the page",
-                  result: { text: "fixture" },
-                },
-              },
-              { type: "response.completed", response: {} },
-            ),
-            { headers: { "content-type": "text/event-stream" } },
-          )
-        }),
-      ),
-    ),
-  ),
-)

packages/ai/test/provider/bedrock-converse.test.ts

@@ -1458,13 +1458,7 @@ describe("Bedrock Converse route", () => {
         expect(headers.get("authorization")).toContain("Credential=AKIACHAINEXAMPLE/")
         expect(headers.get("authorization")).toContain("/ap-southeast-2/bedrock/aws4_request")
       }
-      expect(() => AmazonBedrock.configure({ auth: "sigv4", apiKey: "k" })).toThrow(
-        expect.objectContaining({
-          _tag: "ProviderConfiguration",
-          provider: "amazon-bedrock",
-          message: "Amazon Bedrock SigV4 auth does not accept apiKey",
-        }),
-      )
+      expect(() => AmazonBedrock.configure({ auth: "sigv4", apiKey: "k" })).toThrow("does not accept apiKey")
     }).pipe(
       withProcessEnv({
         ...noAmbientAWS,

packages/ai/test/provider/bedrock-mantle.test.ts

@@ -4,7 +4,7 @@ import { HttpClientRequest } from "effect/unstable/http"
 import { LLM, Message } from "../../src/index.js"
 import { AmazonBedrockMantle } from "../../src/providers.js"
 import { model } from "../../src/providers/amazon-bedrock/mantle.js"
-import { OpenResponses } from "../../src/protocols/open-responses.js"
+import { OpenAIResponses } from "../../src/protocols/openai-responses.js"
 import { compileRequest, LLMClient } from "../../src/route/client.js"
 import { it } from "../lib/effect.js"
 import { withProcessEnv } from "../lib/env.js"
@@ -25,7 +25,7 @@ describe("Amazon Bedrock Mantle provider", () => {
       expect(provider.model).toBe(provider.responses)
       expect(AmazonBedrockMantle.model).toBe(AmazonBedrockMantle.responsesModel)
       expect(model).toBe(AmazonBedrockMantle.responsesModel)
-      expect(provider.model("openai.gpt-oss-120b").route.transport).toBe(OpenResponses.httpTransport)
+      expect(provider.model("openai.gpt-oss-120b").route.transport).toBe(OpenAIResponses.httpTransport)
       const chat = yield* compileRequest(LLM.request({ model: provider.chat("openai.gpt-oss-120b"), prompt: "Hi" }))
       const responses = yield* compileRequest(
         LLM.request({ model: provider.model("openai.gpt-oss-120b"), prompt: "Hi" }),
@@ -38,7 +38,7 @@ describe("Amazon Bedrock Mantle provider", () => {
       })
       expect(responses).toMatchObject({
         route: "bedrock-mantle-responses",
-        protocol: "open-responses",
+        protocol: "openai-responses",
         body: { model: "openai.gpt-oss-120b", store: false },
       })
       expect(provider.model("openai.gpt-oss-120b").route.providerMetadataKey).toBe("mantle")
@@ -178,7 +178,7 @@ describe("Amazon Bedrock Mantle provider", () => {
 const recorded = recordedTests({
   prefix: "bedrock-mantle",
   provider: "amazon-bedrock",
-  protocol: "open-responses",
+  protocol: "openai-responses",
   requires: ["AWS_BEARER_TOKEN_BEDROCK"],
   metadata: { model: "openai.gpt-oss-120b" },
 })

packages/ai/test/provider/conversation-lowering.test.ts

@@ -23,7 +23,7 @@ it.effect("conversation lowering excludes generation settings and tool definitio
       instructions: "Keep the context",
       input: [
         { role: "user", content: [{ type: "input_text", text: "hello" }] },
-        { type: "message", role: "assistant", status: "completed", content: [{ type: "output_text", text: "hi" }] },
+        { type: "message", role: "assistant", content: [{ type: "output_text", text: "hi" }] },
       ],
     })
   }),

packages/ai/test/provider/google-vertex.test.ts

@@ -378,11 +378,7 @@ describe("Google Vertex providers", () => {
 
   test("rejects tuned Gemini models in express mode", () => {
     expect(() => GoogleVertex.configure({ apiKey: "fixture" }).model("endpoints/1234567890")).toThrow(
-      expect.objectContaining({
-        _tag: "ProviderConfiguration",
-        provider: "google-vertex",
-        message: "Google Vertex tuned models do not support Express Mode API keys",
-      }),
+      "Google Vertex tuned models do not support Express Mode API keys",
     )
   })
 })

packages/ai/test/provider/open-responses-errors.test.ts

@@ -1,78 +0,0 @@
-import { expect } from "bun:test"
-import { Effect, Schema } from "effect"
-import { LLM, LLMClient } from "../../src/index.js"
-import { OpenResponses } from "../../src/protocols/open-responses.js"
-import { Meta } from "../../src/providers/index.js"
-import { configure } from "../../src/providers/openai-compatible-responses.js"
-import { it } from "../lib/effect.js"
-import { fixedResponse } from "../lib/http.js"
-import { sseEvents } from "../lib/sse.js"
-
-const decodeEvent = Schema.decodeUnknownEffect(OpenResponses.protocol.stream.event)
-
-it.effect("normalizes flat errors in shared SSE and WebSocket decoding", () =>
-  Effect.gen(function* () {
-    const frame = {
-      type: "error",
-      sequence_number: 4,
-      code: "server_shutting_down",
-      message: "Server is shutting down. Please retry your request.",
-      param: null,
-    }
-    for (const decode of [decodeEvent, OpenResponses.decodeChannelEvent]) {
-      const event = yield* decode(JSON.stringify(frame))
-      expect(event).toEqual({
-        type: "error",
-        sequence_number: 4,
-        error: { code: frame.code, message: frame.message, param: null },
-      })
-
-      for (const unchanged of [
-        event,
-        { type: "error" },
-        {
-          type: "response.failed",
-          response: { id: "resp_failed", error: { code: "server_error", message: "Internal server error" } },
-        },
-        { type: "response.output_text.delta", item_id: "msg_text", delta: "Hello" },
-      ]) {
-        expect(yield* decode(JSON.stringify(unchanged))).toEqual(unchanged)
-      }
-    }
-  }),
-)
-
-it.effect("continues to normalize untyped xAI WebSocket errors", () =>
-  Effect.gen(function* () {
-    const frame = { error: { type: "api_error", message: "gRPC error: Response with id=resp_missing not found" } }
-    expect(yield* OpenResponses.decodeChannelEvent(JSON.stringify(frame))).toEqual({ ...frame, type: "error" })
-  }),
-)
-
-it.effect("retains classification and original error bodies through Meta and generic Responses routes", () =>
-  Effect.gen(function* () {
-    const raw = `{
-  "type": "error",
-  "sequence_number": 4,
-  "code": "server_shutting_down",
-  "message": "Server is shutting down. Please retry your request.",
-  "param": null,
-  "diagnostic": "retain-original-frame"
-}`
-    for (const model of [
-      Meta.configure({ apiKey: "fixture" }).responses("muse-spark-1.3"),
-      configure({ apiKey: "fixture", provider: "gateway", baseURL: "https://responses.example.test/v1" }).model(
-        "example-model",
-      ),
-    ]) {
-      const error = yield* LLMClient.generate(LLM.request({ model, prompt: "Hello" })).pipe(
-        Effect.provide(fixedResponse(sseEvents(raw.replaceAll("\n", "\ndata: ")))),
-        Effect.flip,
-      )
-      expect(error.reason._tag).toBe("ProviderInternal")
-      expect(error.message).toBe("server_shutting_down: Server is shutting down. Please retry your request.")
-      expect(error.reason.body).toBe(raw)
-      expect(error.reason.http?.status).toBe(200)
-    }
-  }),
-)

packages/ai/test/provider/open-responses-replay.test.ts

@@ -1,120 +0,0 @@
-import { describe, expect } from "bun:test"
-import { Effect } from "effect"
-import { LLM, LLMEvent, Message } from "../../src/index.js"
-import { OpenAI } from "../../src/providers.js"
-import { configure } from "../../src/providers/openai-compatible-responses.js"
-import { compileRequest, LLMClient } from "../../src/route/client.js"
-import { it } from "../lib/effect.js"
-import { fixedResponse } from "../lib/http.js"
-import { sseEvents } from "../lib/sse.js"
-
-for (const model of [
-  OpenAI.configure({ apiKey: "test-key" }).responses("example-model"),
-  configure({ apiKey: "test-key", baseURL: "https://responses.example.test/v1" }).model("example-model"),
-]) {
-  describe(`${model.route.protocol} message replay`, () => {
-    const key = model.route.providerMetadataKey ?? "openresponses"
-
-    it.effect("marks assistant text completed regardless of stored status", () =>
-      Effect.gen(function* () {
-        const prepared = yield* compileRequest(
-          LLM.request({
-            model,
-            messages: [
-              ...[undefined, "in_progress", "incomplete", "completed"].map((status, index) =>
-                Message.make({
-                  role: "assistant",
-                  providerMetadata: { [key]: { status } },
-                  content: [
-                    {
-                      type: "text",
-                      text: `Saved ${index}`,
-                      providerMetadata: { [key]: { itemId: `msg_${index}`, phase: "commentary", status } },
-                    },
-                    {
-                      type: "text",
-                      text: `Final ${index}`,
-                      providerMetadata: { [key]: { itemId: `msg_final_${index}`, phase: "final_answer", status } },
-                    },
-                  ],
-                }),
-              ),
-              Message.make({
-                role: "user",
-                content: [{ type: "text", text: "Continue" }],
-                providerMetadata: { [key]: { status: "incomplete" } },
-              }),
-            ],
-          }),
-        )
-        expect(prepared.body.input).toEqual([
-          ...[0, 1, 2, 3].flatMap((index) => [
-            {
-              type: "message",
-              role: "assistant",
-              id: `msg_${index}`,
-              phase: "commentary",
-              status: "completed",
-              content: [{ type: "output_text", text: `Saved ${index}` }],
-            },
-            {
-              type: "message",
-              role: "assistant",
-              id: `msg_final_${index}`,
-              phase: "final_answer",
-              status: "completed",
-              content: [{ type: "output_text", text: `Final ${index}` }],
-            },
-          ]),
-          { role: "user", status: "incomplete", content: [{ type: "input_text", text: "Continue" }] },
-        ])
-      }),
-    )
-
-    it.effect("replays truncated text as completed while retaining the response finish reason", () =>
-      Effect.gen(function* () {
-        const response = yield* LLMClient.generate(LLM.request({ model, prompt: "Respond" })).pipe(
-          Effect.provide(
-            fixedResponse(
-              sseEvents(
-                {
-                  type: "response.output_item.added",
-                  item: { type: "message", id: "msg_partial", status: "in_progress" },
-                },
-                { type: "response.output_text.delta", item_id: "msg_partial", delta: "The next step is" },
-                {
-                  type: "response.output_item.done",
-                  item: {
-                    type: "message",
-                    id: "msg_partial",
-                    status: "incomplete",
-                    content: [{ type: "output_text", text: "The next step is" }],
-                  },
-                },
-                {
-                  type: "response.incomplete",
-                  response: { status: "incomplete", incomplete_details: { reason: "max_output_tokens" } },
-                },
-              ),
-            ),
-          ),
-        )
-        expect(response.finishReason.normalized).toBe("length")
-        expect(response.events.filter(LLMEvent.is.textEnd)).toHaveLength(1)
-        const prepared = yield* compileRequest(
-          LLM.request({ model, messages: [response.message, Message.user("Continue")] }),
-        )
-        expect(prepared.body.input).toEqual([
-          {
-            type: "message",
-            role: "assistant",
-            id: "msg_partial",
-            status: "completed",
-            content: [{ type: "output_text", text: "The next step is" }],
-          },
-          { role: "user", content: [{ type: "input_text", text: "Continue" }] },
-        ])
-      }),
-    )
-  })
-}

packages/ai/test/provider/openai-compatible-responses.test.ts

@@ -91,7 +91,7 @@ describe("Open Responses-compatible route", () => {
       expect(prepared.body.input).toEqual([
         { role: "user", content: [{ type: "input_text", text: "Before." }] },
         { role: "developer", content: "Operator update." },
-        { type: "message", role: "assistant", status: "completed", content: [{ type: "output_text", text: "After." }] },
+        { type: "message", role: "assistant", content: [{ type: "output_text", text: "After." }] },
       ])
     }),
   )
@@ -299,27 +299,23 @@ describe("Open Responses-compatible route", () => {
           type: "message",
           id: "history_1",
           role: "assistant",
-          status: "completed",
           content: [{ type: "output_text", text: "Kept." }],
         },
         {
           type: "message",
           id: `history_${"a".repeat(64)}`,
           role: "assistant",
-          status: "completed",
           content: [{ type: "output_text", text: "Long." }],
         },
         {
           type: "message",
           id: "provider_value/with+symbols",
           role: "assistant",
-          status: "completed",
           content: [{ type: "output_text", text: "Opaque." }],
         },
         {
           type: "message",
           role: "assistant",
-          status: "completed",
           content: [
             { type: "output_text", text: "No suffix." },
             { type: "output_text", text: "No prefix." },
@@ -860,7 +856,6 @@ describe("Open Responses-compatible route", () => {
           type: "message",
           id: "msg_refusal",
           role: "assistant",
-          status: "completed",
           content: [{ type: "output_text", text: "I can't help with that." }],
         },
       ])

packages/ai/test/provider/openai-responses-websocket.recorded.test.ts

@@ -171,7 +171,7 @@ describe("OpenAI Responses WebSocket recorded", () => {
         instructions: "Follow the user's exact reply instruction.",
         input: [
           { role: "user", content: [{ type: "input_text", text: "Reply exactly: Alpha." }] },
-          { role: "assistant", status: "completed", content: [{ type: "output_text", text: "Alpha." }] },
+          { role: "assistant", content: [{ type: "output_text", text: "Alpha." }] },
           { role: "user", content: [{ type: "input_text", text: "Reply exactly: Beta." }] },
         ],
       })
@@ -208,7 +208,7 @@ describe("OpenAI Responses WebSocket recorded", () => {
         instructions: "Follow the user's exact reply instruction.",
         input: [
           { role: "user", content: [{ type: "input_text", text: "Reply exactly: Ready." }] },
-          { role: "assistant", status: "completed", content: [{ type: "output_text", text: "Ready." }] },
+          { role: "assistant", content: [{ type: "output_text", text: "Ready." }] },
           { role: "user", content: [{ type: "input_text", text: "Reply exactly: Recovered." }] },
         ],
       })