wooo/awoooi
1
1
Fork 0
Code Issues 2 Pull Requests 1 Actions 3 Packages Projects Releases Wiki Activity

fix(nvidia): upgrade to meta/llama-3.1-8b-instruct (128k context) to avoid 400 bad request on API
All checks were successful
E2E Health Check / e2e-health (push) Successful in 17s
Details

Browse Source
This commit is contained in:
OG T
2026-03-31 13:49:49 +08:00
parent 11627f25f0
commit 22796c6aff
3 changed files with 8 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
apps/api/models.json
View File

@@ -105,15 +105,15 @@
},
"nvidia": {
"name": "NVIDIA Nemotron (ADR-036)",
"name": "NVIDIA NIM (Llama 3.1 8B / ADR-036)",
"enabled": true,
"priority": 4,
"endpoint": "https://integrate.api.nvidia.com/v1",
"api_path": "/chat/completions",
"models": {
"default": "nvidia/nemotron-mini-4b-instruct",
"tool_calling": "nvidia/nemotron-mini-4b-instruct",
"rca": "nvidia/nemotron-mini-4b-instruct"
"default": "meta/llama-3.1-8b-instruct",
"tool_calling": "meta/llama-3.1-8b-instruct",
"rca": "meta/llama-3.1-8b-instruct"
},
"options": {
"temperature": 0.0,

4
apps/api/src/services/model_registry.py
View File

@@ -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",
}
},
},

4
apps/api/src/services/nvidia_provider.py
View File

@@ -114,8 +114,8 @@ class INvidiaProvider(Protocol):
# NVIDIA NIM API Endpoint
NVIDIA_API_URL = "https://integrate.api.nvidia.com/v1/chat/completions"
# 預設模型 (2026-03-29 ogt: 修正為可用的 mini 模型)
NVIDIA_DEFAULT_MODEL = "nvidia/nemotron-mini-4b-instruct"
# 預設模型 (2026-03-31 ogt: 修正為 128k context 版的 Llama 3.1)
NVIDIA_DEFAULT_MODEL = "meta/llama-3.1-8b-instruct"
# 請求超時 (秒) - Nemotron 延遲 11-45s
NVIDIA_TIMEOUT = 60.0