Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

FreeVideo

FreeVideo: A local inference engine for generating MiniMax H3 videos on consumer GPUs.

What problem it solves

FreeVideo enables local video generation on consumer hardware while coordinating VRAM, system memory, disk, weight placement, compute precision, and attention kernels. It provides a ComfyUI workflow and launchers that simplify setup across Windows, macOS, and Linux.

Key features

  • Hardware-adaptive execution — The engine selects GPU-specific weight formats and probes available attention kernels. It supports int8 on GeForce and RTX 30-series cards and FP8 on workstation and datacenter cards.

    Source: README

  • Low-memory local inference — Weight streaming, asynchronous prefetching, and chunked computation manage memory use. The project documents inference with as little as 8 GB of VRAM and 16 GB of RAM.

    Source: README

  • Multimodal generation inputs — Generation workflows can combine text prompts with first and last frames. Image, video, and audio references are also supported.

    Source: README

  • ComfyUI creative workspace — A dedicated workspace inside ComfyUI supports two-pass sampling, batch generation, and creation history. Node view provides further control for LoRAs and workflow customization.

    Source: README

  • Cross-platform deployment — Launchers for Windows, macOS, and Linux configure ComfyUI, the runtime environment, and models. They can reuse existing models and support offline installation.

    Source: README, setup.sh

Quick start

Copied verbatim from the project's README.

cd ComfyUI/custom_nodes
git clone https://github.com/FlashML-org/FreeVideo.git
git clone https://github.com/FlashML-org/FreeVideo.git && cd FreeVideo
./setup.sh

Who it may suit

This may suit users who want to generate MiniMax H3 videos locally on consumer GPUs, particularly those already using ComfyUI or seeking Windows, macOS, or Linux deployment options.

Topics to discuss

  • How should hardware-adaptive weight placement and attention-kernel selection be evaluated across different consumer GPUs?

  • Which multimodal reference workflows are most useful for local video generation?

  • How can the ComfyUI workspace better support community LoRAs, batch generation, and timeline editing?


Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/FlashML-org/FreeVideo#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.

github.comhttps://github.com/FlashML-org/FreeVideo

Comments

…