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

fix vllm

Browse source
This commit is contained in:
sirius0xdev 2026-05-06 01:14:00 +00:00
parent 882be08f17
commit 7410d84f57
1 changed files with 5 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml
View file

@ -39,11 +39,11 @@ spec:
- --tensor-parallel-size=1
- --tokenizer-mode=hf
- --gpu-memory-utilization=0.98
- --max-model-len=262144
- --max-model-len=187654
- --enable-auto-tool-choice
- --max-num-batched-tokens=65536
- --max-num-seqs=8
- --enable-chunked-prefill
- --max-num-batched-tokens=32768
- --max-num-seqs=16
- --trust-remote-code
- --dtype=auto
- --enable-prefix-caching
@ -59,7 +59,7 @@ spec:
requests:
nvidia.com/gpu: 1
memory: "150Gi"
memory: "120Gi"
cpu: "24"
startupProbe: