model_worker_SDXLTurbo.sh

#!/bin/bash
#SBATCH --job-name=visual_arena_controller
#SBATCH --gres=gpu:a40:1
#SBATCH --qos=m3
#SBATCH --time=50:00:00
#SBATCH -c 4
#SBATCH --qos=deadline
#SBATCH --account=deadline
#SBATCH --mem=10G
#SBATCH --output=log/visual_arena_controller-%j.out
#SBATCH --error=log/visual_arena_controller-%j.err
ipAddr=$(ip addr show $(ip route | awk '/default/ { print $5 }') | grep "inet" | head -n 1 | awk '/inet/ {print $2}' | cut -d'/' -f1)
echo "ipAddr = $ipAddr"

# prepare your environment here
module load cuda-11.7


# put your command here
rm -r /h/tianleli/.cache/huggingface
python3 -m fastchat.serve.model_worker --model-path imagenhub_SDXLTurbo --controller http://172.17.8.150:21001 --port 31009 --worker http://$ipAddr:31009 --host $ipAddr