zhizinan1997/jimeng-free-api-all

What it solves

Jimeng AI Free provides a way to access the image and video generation capabilities of the Jimeng AI platform via a standardized API. It allows users to bypass the official web interface and integrate these models into their own applications, while maximizing the use of free daily credits through account pooling and round-robin token rotation.

How it works

The project acts as an API gateway/proxy that translates OpenAI-compatible requests into calls to Jimeng's internal APIs. It features an account pool that stores encrypted cookies (AES-256-GCM) and automatically rotates through available accounts to distribute the load and manage credits. It also includes a web-based management console for monitoring usage, tracking credits, and managing API keys.

Who it’s for

Developers who want to integrate Jimeng's high-quality image and video generation models into their own tools or AI agents, and users who prefer a centralized management interface for multiple AI accounts.

Highlights

• OpenAI Compatibility: Uses the /v1/chat/completions endpoint, making it easy to integrate with existing AI software.
• Comprehensive Model Support: Supports a wide array of image models (up to 5.0 Lite) and video models (including Seedance series), with automatic synchronization of the latest models from the official site.
• Account Pool Management: Supports multiple cookies with round-robin selection, failure cooldowns, and encrypted storage.
• Visual Management Console: A built-in dashboard for real-time statistics, credit tracking, and a media library to preview generated content.
• Intelligent Generation: Automatically detects aspect ratios from prompts and handles resolution downgrading (e.g., 2K to 1K) if credits are insufficient.
• Multimodal Input: Supports image-to-image and image-to-video generation by passing reference images via the API.