feat: route qwen2.5:3b-instruct to CPU service on cp-2

Model moved from ornith-predictor GPU to dedicated qwen-cpu Deployment on talos-cp-2 (CPU). Frees 1 GPU for ComfyUI. Co-authored-by: poimen <[email protected]>
2026-09-08 18:26:32 -07:00
parent 359645f397
commit d40a0946e1
2 changed files with 2 additions and 2 deletions
@@ -36,7 +36,7 @@ data:
      path: "/v1/chat/completions"
    
    - name: "qwen2.5:3b-instruct"
-      address: "ornith-predictor.llm-serving:80"
+      address: "qwen-cpu.llm-serving:80"
      path: "/v1/chat/completions"
    
    - name: "nomic-ai/nomic-embed-text-v2-moe"