CLIP Image-Text Encoding Service

CLIP Service is used for unified vector encoding of text and images, suitable for cross-modal retrieval, image-text matching, and similarity computation.

Service Information

• Service address:
• Health check address:
• Container image: registry.lazycat.cloud/x/videorag/clip:e5202e3

Standalone Deployment

1. Save the following content to a docker-compose.yml file:

services:
  clip:
    image: registry.lazycat.cloud/x/videorag/clip:e5202e3
    ports:
      - 3000:3000
    environment:
      - AUTO_UNLOAD_INTERVAL=300
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:3000/health"]
      interval: 30s
      timeout: 30s
      retries: 30
      start_period: 30s

2. Run docker-compose up -d in the current directory to start
3. After startup, access http://127.0.0.1:3000/health to verify service health