Verified project record

HKUDS/ViMax

hkuds/vimax is an agentic framework that converts a concept or explicit screenplay into a finished video by orchestrating scriptwriting, storyboarding, character creation, and video generation. It requires external LLM, image, and video generation APIs, and is designed for users who interact via command-line tools and edit configuration files.

Project overview

It consolidates Director, Screenwriter, Producer, and Video Generator functionalities into a single framework to address cross-frame inconsistency and the lack of narrative structure in short AI clips.

Project type: AI Agent · Workflow · Video
Use cases: Video Creation
Deployment: Refer to project documentation
License: MIT

Best for

Creators and developers who need to convert long-form fiction or explicit screenplays into multi-scene, multi-shot videos and are comfortable using CLI tools and editing Python files.

Key capabilities

Converts an explicit screenplay into controllable multi-scene, multi-shot video while preserving its creative intent.
A user interface for the framework is planned but not currently available.

Limitations and risks

The framework aims to overcome the limitation of existing AI tools that typically generate only seconds of footage, though inherent model constraints may still apply.
It requires external API keys for LLM, image, and video generation models, which may incur costs and necessitate a paid service dependency.

Getting started

Setup is rated medium difficulty because it requires configuring external API keys and editing Python files for some workflows. Users clone the repository, install uv and run uv sync, configure API keys in configs/agent.local.yaml, and run vimax tui.

Evidence and sources

GitHub project description: "ViMax: Agentic Video Generation (Director, Screenwriter, Producer, and Video Generator All-in-One)"
README: - ❌ **Limited to Short Clips** - Most AI tools generate only seconds of footage. <br> - ❌ **Consistency Chaos** - Characters and scenes change unpredictably across frames. <br> -…
README: Simply input your concept. ViMax autonomously handles the rest. It orchestrates scriptwriting, storyboarding, character creation, and final video generation—all end-to-end. 🚀
README: ViMax is an agentic video creation framework that connects narrative planning, visual consistency, image generation, video generation, and final assembly in one extensible workflo…
README: ViMax also provides a minimal TUI for interactive agent-based video creation.

Reviewed at: 2026-07-20T14:31:29Z

Open original GitHub project