This project provides a free, OpenAI-compatible API gateway for Jimeng AI's advanced image and video generation models. It targets developers and researchers seeking to integrate powerful AI media creation capabilities into their applications or workflows without direct API costs, offering a built-in management console for monitoring and control.

How It Works

This service acts as a reverse proxy, leveraging reverse-engineered APIs from the official Jimeng AI platform. It aggregates multiple image (4.x/3.x) and video generation models, authenticates requests using user-provided Jimeng Session IDs (supporting round-robin across multiple accounts), and exposes a standardized /v1/chat/completions endpoint compatible with OpenAI's API format. A Dockerized deployment simplifies setup, while an integrated web console offers real-time monitoring, media management, and credit tracking.

Quick Start & Requirements

• Primary install / run command (Docker):

  docker run -it -d --init --name jimeng-free-api \
    -p 8001:8000 \
    -v jimeng-data:/app/data \
    -e TZ=Asia/Shanghai \
    ghcr.io/zhizinan1997/jimeng-free-api-all:latest
  

  Access the management console at http://localhost:8001.
• Local Development: Requires Node.js (v16+). Clone the repository, run npm install, npm run build, and npm start. The service will run at http://localhost:8000.
• Non-default prerequisites: A Jimeng official website account is required to obtain Session IDs for authentication.
• Estimated setup time: Minimal via Docker; local setup depends on Node.js environment. Docker image size is approximately 280MB.
• Links: Base project: LLM-Red-Team/jimeng-free-api.

Highlighted Details

• Extensive Model Support: Integrates 5 image models (4.5, 4.1, 4.0, 3.1, 3.0, 2.0-pro) and 5 video models (3.0 Pro, 3.0, 3.0 Fast, S2.0, 2.0 Pro).
• OpenAI Compatibility: Provides a /v1/chat/completions endpoint, supporting multimodal inputs (text, image URLs) for seamless integration with existing AI applications.
• Free Tier Utilization: Maximizes official daily free credits (66 per account) through multi-token round-robin scheduling.
• Integrated Management Console: Features a web UI for real-time statistics, API key status, credit tracking, log viewing, and a media library for generated assets.
• Smart Generation Features: Supports image-to-image, image-to-video (with start/end frames), automatic resolution scaling (2K/1K), intelligent aspect ratio detection, and duration adjustment based on keywords.

Maintenance & Community

This project is a secondary development based on the LLM-Red-Team/jimeng-free-api repository. Its maintenance and stability are inherently tied to the reverse-engineered nature of the underlying Jimeng APIs, which are not officially supported. No specific community channels (e.g., Discord, Slack) are listed.

Licensing & Compatibility

Released under the MIT License. However, the project explicitly states it is for personal research, exchange, and learning purposes only and is not intended for commercial use. Compatibility with closed-source applications is possible via the OpenAI-compatible API, but commercial deployment is restricted by the project's usage terms.

Limitations & Caveats

The service's stability is not guaranteed due to its reliance on reverse-engineered APIs, which may change or become unavailable without notice. A known issue exists where 4.x models requesting 2K resolution might be rejected by the Jimeng server (error code 1006), leading to an automatic downgrade to 1K resolution. The project is strictly limited to non-commercial use.