mirror of
https://github.com/bytedance/deer-flow.git
synced 2026-06-10 09:25:57 +00:00
AochenShen99
4093c83383
* Share assistant payload replay matching * fix(provider): recover assistant field when ordinal AI index is taken The mismatch-length fallback in `_match_ai_message` only tried the exact `fallback_ordinal` AI index. When serialization drops or reorders an assistant message, a unique signature match can consume a non-ordinal index, leaving a later ambiguous payload's ordinal already used — so its provider field (e.g. `reasoning_content`) was silently dropped. Scan forward from the ordinal for the next unused `AIMessage` (wrapping to earlier indices) to preserve the positional bias while still recovering the field. Forward scanning avoids a naive min-unused pick that could restore the wrong field after a leading message is dropped. Add a regression test for the dropped-leading-message case. * fix(provider): avoid earlier assistant fallback replay
60 lines
2.1 KiB
Python
60 lines
2.1 KiB
Python
"""Patched ChatDeepSeek that preserves reasoning_content in multi-turn conversations.
|
|
|
|
This module provides a patched version of ChatDeepSeek that properly handles
|
|
reasoning_content when sending messages back to the API. The original implementation
|
|
stores reasoning_content in additional_kwargs but doesn't include it when making
|
|
subsequent API calls, which causes errors with APIs that require reasoning_content
|
|
on all assistant messages when thinking mode is enabled.
|
|
"""
|
|
|
|
from typing import Any
|
|
|
|
from langchain_core.language_models import LanguageModelInput
|
|
from langchain_deepseek import ChatDeepSeek
|
|
|
|
from deerflow.models.assistant_payload_replay import restore_assistant_payloads, restore_reasoning_content
|
|
|
|
|
|
class PatchedChatDeepSeek(ChatDeepSeek):
|
|
"""ChatDeepSeek with proper reasoning_content preservation.
|
|
|
|
When using thinking/reasoning enabled models, the API expects reasoning_content
|
|
to be present on ALL assistant messages in multi-turn conversations. This patched
|
|
version ensures reasoning_content from additional_kwargs is included in the
|
|
request payload.
|
|
"""
|
|
|
|
@classmethod
|
|
def is_lc_serializable(cls) -> bool:
|
|
return True
|
|
|
|
@property
|
|
def lc_secrets(self) -> dict[str, str]:
|
|
return {"api_key": "DEEPSEEK_API_KEY", "openai_api_key": "DEEPSEEK_API_KEY"}
|
|
|
|
def _get_request_payload(
|
|
self,
|
|
input_: LanguageModelInput,
|
|
*,
|
|
stop: list[str] | None = None,
|
|
**kwargs: Any,
|
|
) -> dict:
|
|
"""Get request payload with reasoning_content preserved.
|
|
|
|
Overrides the parent method to inject reasoning_content from
|
|
additional_kwargs into assistant messages in the payload.
|
|
"""
|
|
# Get the original messages before conversion
|
|
original_messages = self._convert_input(input_).to_messages()
|
|
|
|
# Call parent to get the base payload
|
|
payload = super()._get_request_payload(input_, stop=stop, **kwargs)
|
|
|
|
restore_assistant_payloads(
|
|
payload.get("messages", []),
|
|
original_messages,
|
|
restore_reasoning_content,
|
|
)
|
|
|
|
return payload
|