Apppublic

weekev/babylon-chat

sourceHugging Faceupdated 3y agoView on Hugging Face

0likes

rootdocker-compose.gpu.yaml

docker-compose.gpu.yaml24 linesDownload Raw Back to root

1version: "3.8"2 3services:4  gpt4all_gpu:5    image: ghcr.io/huggingface/text-generation-inference:0.9.36    container_name: gpt4all_gpu7    restart: always #restart on error (usually code compilation from save during bad state)8    environment:9      - HUGGING_FACE_HUB_TOKEN=token10      - USE_FLASH_ATTENTION=false11      - MODEL_ID=''12      - NUM_SHARD=113    command: --model-id $MODEL_ID --num-shard $NUM_SHARD14    volumes:15      - ./:/data16    ports:17      - "8080:80"18    shm_size: 1g19    deploy:20      resources:21        reservations:22          devices:23            - driver: nvidia24              capabilities: [gpu]