feat: update VLLM service configuration in docker-compose

2026-07-21 02:06:47 +00:00 · 2026-02-12 13:48:23 +08:00
parent 78cfb8058c
commit 451e7f9566
1 changed files with 7 additions and 2 deletions
@@ -3,9 +3,14 @@ services:
    image: docker.io/vllm/vllm-openai:cu130-nightly
    command:
      - --model
-      - Qwen/Qwen3-0.6B
+      # - Qwen/Qwen3-4B-Instruct-2507
+      - Qwen/Qwen3-VL-2B-Instruct
+      - --gpu-memory-utilization
+      - "0.75"
+      - --max-model-len
+      - "2048"
    ports:
-      - "8000:8000"
+      - "9000:8000"
    environment:
      HF_TOKEN: ${HF_TOKEN}
    volumes: