fix(nvidia): upgrade to meta/llama-3.1-8b-instruct (128k context) to avoid 400 bad request on API

2026-03-31 13:49:49 +08:00
parent 11627f25f0
commit 22796c6aff
3 changed files with 8 additions and 8 deletions
--- a/apps/api/src/services/model_registry.py
+++ b/apps/api/src/services/model_registry.py
@@ -144,8 +144,8 @@ class ModelRegistry:
                # 2026-03-29 ogt: P2-3 加入 NVIDIA (ADR-036)
                "nvidia": {
                    "models": {
-                        "default": "nvidia/nemotron-mini-4b-instruct",
-                        "tool_calling": "nvidia/nemotron-mini-4b-instruct",
+                        "default": "meta/llama-3.1-8b-instruct",
+                        "tool_calling": "meta/llama-3.1-8b-instruct",
                    }
                },
            },