#3737th of 120 in Image and videoFor image, video and design, video and animation

freevideo

Local MiniMax H3 video generation on consumer GPUs, run as a ComfyUI plugin

Releases and problems feed (RSS)

Stars
4k
License
Apache-2.0
Last commit
Nov 2026
Last release
Nov 2026
Language
Python
Stars, last 9 weeks+3,549
17 Sept 2026: 441 stars24 Sept 2026: 441 stars1 Oct 2026: 441 stars8 Oct 2026: 1,831 stars15 Oct 2026: 2,402 stars22 Oct 2026: 2,678 stars29 Oct 2026: 2,899 stars5 Nov 2026: 3,234 stars12 Nov 2026: 3,482 stars19 Nov 2026: 3,990 stars

441 on 17 Sept 20263,990 on 19 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

FreeVideo is a local inference engine for the MiniMax H3 video model, using the 8-step VDN-H3 weights from OpenVDN. It adapts weight format (int8 or FP8), memory streaming and attention kernels to the GPU, and runs inside ComfyUI with launchers for Windows, macOS and Linux. Inputs include text prompts, first and last frames, and image, video or audio references.

Who it is for: Users generating AI video locally on consumer GPUs through ComfyUI

Strengths

  • Runs MiniMax H3 with as little as 8 GB VRAM and 16 GB RAM
  • Picks int8 or FP8 weights per GPU and probes available attention kernels
  • Launchers for Windows, macOS and Linux install ComfyUI and models, with offline install
  • Supports text, first/last frame, and image, video and audio references plus community LoRAs

Weaknesses

  • Model weights use the MiniMax H3 Community License with territorial and acceptable-use limits
  • macOS support is an Apple silicon preview, tested only on an M5 with 24 GB
  • Mac build is not notarized, so macOS blocks first launch until approved manually
  • Tied to ComfyUI and one model family (MiniMax H3 / VDN-H3)

What it needs

  • RAM ≥ 16 GB
  • GPU required
  • Needs ComfyUI
  • Models: MiniMax H3, VDN-H3, MiniMax H3 LoRAs

What changed

From the release notes. No flag means the note does not say so, not that the upgrade is safe.

Known problems

Open now, reported in the last 90 days

Titles are the reporters' own words, taken from the project's GitHub issues and ranked by reactions. We have not tried to reproduce any of them.

Also in Image and video

See all 120
Also in Image and video
RankProjectScore
1frigateLocal NVR with realtime object detection for IP cameras37.8k stars+950 (+2.6%) in 30 daysMIT94 out of 100
2toonflow-appInfinite-canvas tool for AI short drama and comic video production22.4k stars+3,446 (+18%) in 30 daysMIT89 out of 100
3MoneyPrinterTurboGenerates short videos from a topic, with script, footage, voiceover and subtitles144.9k starsMIT88 out of 100
#4InvokeAIWeb UI for diffusion image generation with inpainting canvas and node workflows29.1k starsApache-2.088 out of 100
#5autoclipTurns a video link into short clips with covers and captions12.5k stars+2,045 (+20%) in 30 daysMIT88 out of 100