fix vllm errors
This commit is contained in:
parent
755bdb39ba
commit
197212aacb
1 changed files with 6 additions and 7 deletions
|
|
@ -34,14 +34,13 @@ spec:
|
||||||
args:
|
args:
|
||||||
- "--model"
|
- "--model"
|
||||||
- "Jiunsong/supergemma4-26b-abliterated-multimodal" # Uncensored Abliterated Gemma 26B (Native)
|
- "Jiunsong/supergemma4-26b-abliterated-multimodal" # Uncensored Abliterated Gemma 26B (Native)
|
||||||
- "--dtype"
|
- --dtype=auto
|
||||||
- "bfloat16"
|
- "bfloat16"
|
||||||
- "--max-model-len"
|
- --max-model-len=8192
|
||||||
- "8192"
|
- --trust-remote-code
|
||||||
- "--trust-remote-code"
|
- --gpu-memory-utilization=0.80
|
||||||
- "--gpu-memory-utilization"
|
- --enforce-eager
|
||||||
- "0.95"
|
|
||||||
- "--enforce-eager"
|
|
||||||
ports:
|
ports:
|
||||||
- containerPort: 8000
|
- containerPort: 8000
|
||||||
resources:
|
resources:
|
||||||
|
|
|
||||||
Loading…
Add table
Reference in a new issue