sglang MODEL=/home/fuhua/.cache/modelscope/hub/models/Qwen/Qwen3___8-27B # MODEL...

sglang MODEL=/home/fuhua/.cache/modelscope/hub/models/Qwen/Qwen3___8-27B # MODEL=/home/fuhua/.cache/modelscope/hub/models/Qwen/Qwen3___8-27B-FP8 exec /home/fuhua/anaconda3/envs/max395/bin/python -m sglang.launch_server \ --model-path "$MODEL" \ --served-model-name Qwen3.8-27B \ --host 0.0.0.0 --port 30000 \ --mem-fraction-static 0.94 \ --context-length 262144 \ --max-running-requests 8 \ --chunked-prefill-size 2048 \ --attention-backend triton \ --disable-cuda-graph \ --max-total-tokens 262144 \ --max-mamba-cache-size 8 \ --tool-call-parser qwen3_coder \ --speculative-algorithm NEXTN \ --speculative-num-steps 3 \ --speculative-eagle-topk 1 \ --speculative-num-draft-tokens 4