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

Update rtx6000-vllm.yaml

Browse source
This commit is contained in:
sirius0xdev 2026-05-22 23:27:18 -04:00 • committed by GitHub
parent 6df297086c
commit 54267b0818
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 4 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml
View file

@ -37,14 +37,14 @@ spec:
- --port=8000 - --port=8000
- --tensor-parallel-size=1 - --tensor-parallel-size=1
- --tokenizer-mode=hf - --tokenizer-mode=hf
- --gpu-memory-utilization=0.90 - --gpu-memory-utilization=0.90
#- --enforce-eager
- --max-model-len=131072 - --max-model-len=131072
- --enable-auto-tool-choice - --enable-auto-tool-choice
- --kv-cache-dtype=bfloat16 - --kv-cache-dtype=bfloat16
- --max-num-batched-tokens=32768 - --max-num-batched-tokens=65536
- --max-num-seqs=32 - --max-num-seqs=64
- --block-size=32 - --block-size=32
- --language-model-only
- --enable-chunked-prefill - --enable-chunked-prefill
- --trust-remote-code - --trust-remote-code
- --dtype=auto - --dtype=auto