textgen Docker & Self-Hosted Deployment

GitHub

Open-source desktop app for local LLMs. Text, vision, tool-calling, OpenAI/Anthropic-compatible API.

46,960 stars Python Python 1 file · ~17,975 tokens
RAW Compose
1. Pull Image (DOCKERHUB) docker pull oobabooga/textgen
2. Docker Run (1-Click Launch) docker run -d --name textgen --restart always oobabooga/textgen:latest
3. Podman Run podman run -d --name textgen --restart always oobabooga/textgen:latest
4. Skopeo Inspect (Zero-Pull Manifest) skopeo inspect docker://docker.io/oobabooga/textgen:latest
5. ORAS Pull (OCI Artifact) oras pull oobabooga/textgen:latest
6. Compose Deployment docker compose up -d
docker/amd/docker-compose.yml
Download .yml 
version: "3.3"
services:
  textgen:
    build:
      context: .
      args:
        BUILD_EXTENSIONS: ${BUILD_EXTENSIONS:-}
        APP_GID: ${APP_GID:-6972}
        APP_UID: ${APP_UID:-6972}
    env_file: .env
    user: "${APP_RUNTIME_UID:-6972}:${APP_RUNTIME_GID:-6972}"
    ports:
      - "${HOST_PORT:-7860}:${CONTAINER_PORT:-7860}"
      - "${HOST_API_PORT:-5000}:${CONTAINER_API_PORT:-5000}"
    stdin_open: true
    group_add:
      - video
    tty: true
    ipc: host
    devices:
      - /dev/kfd
      - /dev/dri 
    cap_add: 
      - SYS_PTRACE
    security_opt:
      - seccomp=unconfined
    volumes:
      - ./user_data:/home/app/textgen/user_data