Update rtx6000-vllm.yaml
This commit is contained in:
parent
08e9998019
commit
864cfd2cc9
1 changed files with 2 additions and 2 deletions
|
|
@ -40,7 +40,7 @@ spec:
|
||||||
- --tokenizer-mode=hf
|
- --tokenizer-mode=hf
|
||||||
- --gpu-memory-utilization=0.95
|
- --gpu-memory-utilization=0.95
|
||||||
- --kv-cache-dtype=fp8_e5m2
|
- --kv-cache-dtype=fp8_e5m2
|
||||||
- --max-model-len=131072
|
- --max-model-len=256389
|
||||||
- --enable-auto-tool-choice
|
- --enable-auto-tool-choice
|
||||||
- --enable-chunked-prefill
|
- --enable-chunked-prefill
|
||||||
- --max-num-batched-tokens=8192
|
- --max-num-batched-tokens=8192
|
||||||
|
|
@ -48,7 +48,7 @@ spec:
|
||||||
- --trust-remote-code
|
- --trust-remote-code
|
||||||
- --dtype=auto
|
- --dtype=auto
|
||||||
- --enable-prefix-caching
|
- --enable-prefix-caching
|
||||||
- --tool-call-parser=qwen3_xml
|
- --tool-call-parser=qwen3_coder
|
||||||
- --reasoning-parser=qwen3
|
- --reasoning-parser=qwen3
|
||||||
- --disable-custom-all-reduce
|
- --disable-custom-all-reduce
|
||||||
ports:
|
ports:
|
||||||
|
|
|
||||||
Loading…
Add table
Reference in a new issue