# antirez/h3.c

> Native MiniMax-H3 inference engine for Apple Silicon on Metal: generates video and audio from text prompts, supports first/last-frame conditioning and Ref2VA references.

- Magnitude: 1.3 out of 10 — Steady
- Stars: 2,971 total · +0 stars measured 2026-10-27, 09:12–11:09 UTC, ≈ 3 by evening
- Star trust: star growth looks organic
- Category: Generative media · Language: C · License: MIT · Created: 2026-08-09 · Last push: 2026-08-11
- GitHub: https://github.com/antirez/h3.c · Page: https://gitnova.dev/en/r/antirez/h3.c

## Useful for

- Run an interactive session to generate video from a text prompt
- Generate a short video with specified first and last frames
- Reduce memory usage during generation with --ssd-streaming

## Why it’s here

- Star-counter measurements on 2026-10-27 (UTC), 09:12–11:09: 2971 → 2971 stars (+0). This is the change over that interval.
- Estimated end-of-day forecast: about +3 stars, using observed gains and the previous day.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 249
- Issues and pull requests: 81
- Watchers: 26
- Average over the last week: 5 per day
- Usual pace: 8 per day
- Stars in the last hour (measured): 0

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-09-28 … 2026-10-27: 5, 9, 5, 4, 1, 11, 4, 4, 9, 17, 13, 5, 7, 6, 7, 10, 10, 3, 6, 12, 8, 7, 3, 5, 5, 6, 3, 7, 9, 0

## Similar by description

1. **MiniMax-AI/MiniMax-H3** — 3.4 · Breakout · Generative media · Python · +61 stars measured 2026-10-27, 00:36–10:53 UTC, ≈ 120 by evening
   MiniMax H3 is an omni-modal generative system that understands text, images, video and audio and generates video with native stereo audio up to 2K and 15 seconds. The repository bundles prompt-writing and style-specific video generation…
   Full card: https://gitnova.dev/en/r/MiniMax-AI/MiniMax-H3.md
2. **Enderfga/SolarRenderer** — 1.7 · Steady · Generative media · Python · +9 stars measured 2026-10-27, 00:41–11:00 UTC, ≈ 20 by evening
   A generative renderer that turns a structure video (grey box, semantic maps, skeletons, depth, mesh, sketches) into a finished clip in four steps on MiniMax-H3 with two LoRAs. The look is set by the first frame and a prompt.
   Full card: https://gitnova.dev/en/r/Enderfga/SolarRenderer.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-27 11:09 UTC, updated every 30 minutes.