vllm
serve
/opt/local-models/qwen36-35b-a3b-f16-gguf/Qwen3.6-35B-A3B-F16.gguf
--served-model-name
Qwen3.6-35B-A3B-F16-GGUF-fusedgdn-base-rmsnorm
--enable-auto-tool-choice
--tool-call-parser
hermes
--dtype
half
--host
0.0.0.0
--port
8001
--tensor-parallel-size
4
--max-model-len
131072
--gpu-memory-utilization
0.95
--trust-remote-code
--generation-config
vllm
-O=3
--reasoning-parser
qwen3
--async-scheduling
--load-format
gguf
--tokenizer
/root/.cache/huggingface/hub/models--Qwen--Qwen3.6-35B-A3B/snapshots/995ad96eacd98c81ed38be0c5b274b04031597b0
--hf-config-path
/opt/vllm_patch_bundle/config_text
--language-model-only
