wehub-resource-sync
7 lines
209 B
Bash
7 lines
209 B
Bash
MODEL_PATH="[MODEL_PATH]"
|
|
python -m vllm.entrypoints.openai.api_server \
|
|
--model $MODEL_PATH \
|
|
--port 8000 \
|
|
--max-model-len 131072 \
|
|
--tensor-parallel-size 4 \
|
|
--gpu-memory-utilization 0.8 |