jiulaikankan/

CAPF-Qwen3-4B-Thinking-Search

vllm serve jiulaikankan/CAPF-Qwen3-4B-Thinking-Search \
    --port 8009 \
    --served-model-name qwen3 \
    --async-scheduling \
    --data-parallel-size 1 \
    --tensor-parallel-size 1 \
    --gpu-memory-utilization 0.9 \
    --enable-auto-tool-choice --tool-call-parser hermes \
    --reasoning-parser deepseek_r1 \
    --max-num-seqs 100 \
    --enforce-eager