You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
38 lines
1.0 KiB
38 lines
1.0 KiB
version: '3.8'
|
|
|
|
services:
|
|
rezero:
|
|
build:
|
|
context: .
|
|
dockerfile: Dockerfile
|
|
image: rezero
|
|
container_name: rezero
|
|
volumes:
|
|
- .:/app
|
|
# Windows-compatible path for huggingface cache
|
|
- ${USERPROFILE}/.cache/huggingface:/root/.cache/huggingface
|
|
ports:
|
|
- "7860:7860" # Gradio web interface
|
|
env_file:
|
|
- .env
|
|
# Windows with Docker Desktop + WSL2 uses this format
|
|
environment:
|
|
- NVIDIA_VISIBLE_DEVICES=all
|
|
- GRADIO_SERVER_NAME=0.0.0.0
|
|
- GRADIO_SERVER_PORT=7860
|
|
- GRADIO_SHARE=False
|
|
- GENERATOR_MODEL_DIR=Menlo/ReZero-v0.1-llama-3.2-3b-it-grpo-250404
|
|
# Use runtime instead of deploy for Windows Docker Desktop
|
|
runtime: nvidia
|
|
# Create model directory and run the app
|
|
command: >
|
|
bash -c "mkdir -p /app/models/generator &&
|
|
python app.py"
|
|
restart: unless-stopped
|
|
shm_size: 8gb
|
|
healthcheck:
|
|
test: ["CMD", "curl", "-f", "http://localhost:7860/"]
|
|
interval: 30s
|
|
timeout: 10s
|
|
retries: 3
|
|
start_period: 40s |