Noesion Radarnoesion.ai
Release2026-08-27

fal released H3 Max, a retuned version of the open-weights MiniMax H3 video model that it says produces a five-second clip in about three seconds. In fal's own human preference tests, and in benchmarks from Artificial Analysis and Design Arena, it ranked first for overall quality, prompt following and look against twelve other video models. It is available now through fal's playground, agent and API, at half price for the first week.

What changed

The original MiniMax H3 endpoint produced comparable video far more slowly.

What it unlocks

Generating short video clips faster than they play, making high-volume or interactive video generation practical.

  • 5-second video in under 3 seconds
  • ~35x throughput of official H3 endpoint
  • 50% off for first week
  • 12 rival video models compared

What you need to act on it

  • a fal account
  • API or Playground access

Sources

Send this to someone who needs it

Shares the story and its sources. Nothing about you.

What does this mean for your job?

This is the story as everyone gets it. Once a week we send you the version written for your role — what changed, why it matters for the work you actually do, and one thing to try. Free while we tune it.