MiniMax H3 Max Live Reaches Twitch as AI Video Streams Keep Generating in Real Time

News Editor

2026-08-31 07:28:17

fal engineer Rehan Sheikh has connected H3 Max directly to Twitch in an experiment that turns AI video generation into a live, continuously updating stream. Because the model can generate footage faster than viewers consume it, the system can play one segment while rendering the next in the background, making a theoretically endless broadcast possible. Sheikh described the setup as an "infinite interdimensional television," and MiniMax later reposted the experiment. fal then continued with a broader H3 Max Live test. The stream is not pre-rendered: viewers can type !prompt in the Twitch chat, and fal said new instructions can appear on screen within seconds, letting the audience shape each scene as the broadcast unfolds. According to the source material, the lack of a fixed script or running time means the program can continue as long as generation continues. The report said the key shift is speed. H3 Max has reduced waiting time to below playback time, clearing a threshold that previously kept AI video from functioning as a true live format. fal Research said H3 Max was trained on top of MiniMax’s open-source H3 and accelerated for fal’s inference system. Official figures cited in the report said it can generate a 5-second video in under 3 seconds, with throughput about 35 times that of MiniMax’s official H3 API. On Artificial Analysis’ audio-enabled ranking, its image-to-video model sits first at 1202 Elo, while its text-to-video model ranks third at 1235 Elo behind only Wan 3.0 and Gemini Omni Flash.

fal engineer Rehan Sheikh has plugged H3 Max directly into Twitch, turning an AI video model into a live stream that can keep generating as it airs.

According to the BlockBeats brief, the model produces video faster than viewers can watch it. That lets the stream play the current segment while the next one is rendered in the background, making a theoretically endless broadcast possible. Sheikh called the format "infinite interdimensional television." MiniMax later reposted the experiment.

Chat prompts can change the stream within seconds

fal went on to run a broader H3 Max Live test. The content in the stream is not generated in advance. Viewers can type !prompt directly into the Twitch chat, and fal said new instructions can show up on screen within seconds, allowing each scene to be influenced by the audience.

The stream has no fixed script and no set duration. As long as generation continues, the show can continue as well.

Speed clears the biggest obstacle to true live AI video

The report said generation speed had been the biggest barrier to this kind of format. If a 15-second clip is followed by a wait of several minutes for the next one, the result is not a real live stream. H3 Max has now pushed that wait below playback time, so the AI can prepare the next segment while viewers are still watching the previous one.

In the source description, that marks a shift for video models from a wait-for-output workflow to one that can continuously supply real-time content.

Model background and performance figures

H3 Max is a version trained by fal Research on top of MiniMax’s open-source H3, with acceleration tuned for fal’s inference system.

Official figures cited in the report said the model can generate a 5-second video in under 3 seconds. Throughput is about 35 times that of MiniMax’s official H3 interface.

Artificial Analysis rankings

On Artificial Analysis’ audio-enabled leaderboard, H3 Max ranks first in image-to-video at 1202 Elo. In text-to-video, it ranks third at 1235 Elo, behind only Wan 3.0 and Gemini Omni Flash.

This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan.

Disclaimer:

The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.

Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.