Skip to content

DEV.md

Latest commit

 

History

History
28 lines (19 loc) · 925 Bytes

DEV.md

File metadata and controls

28 lines (19 loc) · 925 Bytes
Raw

AccelBrain Model Server Docker (Based on Ollama)

This project is forked from ollama/ollama,
and is customized to build a Dockerized model server for use in the AccelBrain system.

🛠️ Build Workflow

  1. Prepare two self-hosted GitHub runners:

    • One with x86_64 (amd64) architecture and GPU support
    • One with ARM64 (arm) architecture and GPU support

  2. Update versioning information:

    • Edit CHANGELOG.md
    • Modify version/version.json with the new version info

  3. Trigger the GitHub Actions build-release workflow:

    • Push a tag (e.g., v1.2.3) to start the workflow
    • Multi-architecture Docker images will be built and published automatically

📦 Docker Images

  • Once built, images will be published to Docker Hub: 
    docker pull innodiskorg/ollama:<version>