GitHub radar
One Client for Seedance, Kling, MiniMax and Wan Video AI
A unified Python client and web UI for generating AI video via Seedance, Kling, MiniMax/Hailuo, and Wan — all from one interface, no Node.js required.
An async Python library and web UI that brings four major video generation services under one roof: Seedance (ByteDance), Kling (Kuaishou), MiniMax/Hailuo, and Wan (Alibaba). Supports text-to-video and image-to-video, live task status polling, and MP4 download. The web interface is bundled directly into the Python package — no Node.js or frontend build step needed. Notable model specs from the README: Seedance 2.5 generates up to 30 seconds of video with native audio, MiniMax H3 delivers 4K at 60fps for up to 30 seconds, Kling 3.0 supports multi-shot with native audio, Wan 3.0 covers up to 1080p for 30 seconds. There is also an MCP server included for agent integration.
Why a vibe-coder should care
Instead of switching between four different dashboards to compare providers and find the best current model, you work from one web UI. Your agent can set it up once — then you pick the provider and model from a dropdown and generate.
How to install
Copy this and send it to your agent — Claude Code, Codex, any of them:
Deploy the video generation client from https://github.com/letorig/video-generator-client — follow the README, create .env from the example, start the web server and open it in the browser. Ask me if you need API keys for specific providers.
Runs on a regular laptop — only Python needed. Each video provider requires its own paid API key.
Open on GitHub▌ More finds