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

Fix A100 crashloop by cleaning up invalid vllm args

Browse source
This commit is contained in:
Sirius Claw 2026-04-20 23:45:43 +00:00
parent e1cf661692
commit 29f57b48a0
1 changed files with 2 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
apps/base/customer1/openclaw/vllm-gemma.yaml
View file

@ -36,16 +36,12 @@ spec:
- --host=0.0.0.0 - --host=0.0.0.0
- --port=8000 - --port=8000
- --tensor-parallel-size=1 - --tensor-parallel-size=1
- --gpu-memory-utilization=0.95 # ← Critical: lower than default 0.90 - --gpu-memory-utilization=0.95
- --max-model-len=24576 # Start conservative (you can raise to 32768 later) - --max-model-len=24576
- --enable-auto-tool-choice - --enable-auto-tool-choice
- --trust-remote-code - --trust-remote-code
- --dtype=auto - --dtype=auto
- --tool-call-parser=pythonic
- --reasoning-parser=qwen3
- --enable-prefix-caching - --enable-prefix-caching
- --enable-chunked-prefill
- --kv-cache-dtype=fp8
##- --limit-mm-per-prompt='{"image"=1, "video"=0}' # Optional: limit vision if not heavily using images yet ##- --limit-mm-per-prompt='{"image"=1, "video"=0}' # Optional: limit vision if not heavily using images yet
ports: ports:
- containerPort: 8000 - containerPort: 8000