sirius/gcloud-lab
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix: correct huggingface model identifier for dolphin 32b and remove awq requirement for a100

Browse source
This commit is contained in:
SiriusClaw 2026-04-18 17:22:37 +00:00
parent 9967b4b39d
commit 1623ccb6e5
1 changed files with 1 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
apps/base/customer1/openclaw/vllm-gemma.yaml
View file

@ -28,9 +28,7 @@ spec:
command: ["python3", "-m", "vllm.entrypoints.openai.api_server"] command: ["python3", "-m", "vllm.entrypoints.openai.api_server"]
args: args:
- "--model" - "--model"
- "cognitivecomputations/dolphin-2.9.4-qwen2.5-32b-AWQ" # Uncensored 32B Reasoning Heavyweight (Quantized) - "cognitivecomputations/dolphin-2.9.2-qwen1.5-32b" # Native 32B Uncensored (fits perfectly in 80GB VRAM)
- "--quantization"
- "awq"
- "--dtype" - "--dtype"
- "half" - "half"
- "--max-model-len" - "--max-model-len"