File size: 1,019 Bytes
9f3bc09 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 |
#!/bin/bash
set -euo pipefail
CONDA_ROOT="/mnt/hwfile/zhangfan.p/miniconda3"
ENV_PATH="$CONDA_ROOT/envs/llamafac"
export PATH="$ENV_PATH/bin:$PATH"
read -p "Enter config (0 for fine-tuned, 1 for base): " CONFIG
export CONFIG=${CONFIG:-0}
if [ $CONFIG -eq 0 ]; then
export CUDA_VISIBLE_DEVICES=1
export MODEL_NAME="ea-dev/eval-agent-vbench-base-table"
export PORT=12333
export GPU_MEMORY_UTILIZATION=0.7
else
export CUDA_VISIBLE_DEVICES=2
export MODEL_NAME="qwen/Qwen2.5-3B-Instruct"
export PORT=12334
export GPU_MEMORY_UTILIZATION=0.7
fi
# Launch using vllm serve command
echo "Starting Qwen2.5-3B eval agent server on 0.0.0.0:${PORT}..."
echo "Model: ${MODEL_NAME}"
echo "GPU Memory Utilization: ${GPU_MEMORY_UTILIZATION}"
exec python -u -m vllm.entrypoints.openai.api_server ${MODEL_NAME} \
--host 0.0.0.0 \
--port ${PORT} \
--gpu-memory-utilization ${GPU_MEMORY_UTILIZATION} \
--trust-remote-code \
--max-model-len 16384 \
--served-model-name eval-agent |