CLASS A · FASTER CLIPOFFICIAL CLAIM
H3 Max
BY FAL
A fal post-trained MiniMax H3 variant built for high-quality, faster-than-playback clip generation.
Open fal launch announcementREALTIME VERDICT
CLASS AThe best recorded value is RTR 0.60, scoped to inference timing and labeled as an official claim.
LAST VERIFIED · 2026-09-03BEST RTR0.60inference timing
OUTPUT480p / 768pFPS not reported
DURATION5–15 secondsQueued API
ACCESSAPI$0.08 / output second at 768p
WHAT WE KNOW
Capabilities with evidence attached.
01
fal reports a 5-second clip in under 3 seconds.
02
Native audio is generated with the video.
03
Text, image, and reference-conditioned endpoints are available.
BEST FIT
Next-scene generationAI TVInteractive stories
EVIDENCE GAP
Can you reproduce this model?
We are looking for endpoint, region, queue, first-frame, and last-byte timing across repeated runs. Raw JSON is more valuable than a screenshot.
Contribute a run