H3 Max Turbo arrives on fal
A higher-throughput H3 Max endpoint exposes backend inference timing and two prompt expansion modes.
LAUNCH RADAR
A dated feed of launches and changes that can alter what developers are able to ship.
A higher-throughput H3 Max endpoint exposes backend inference timing and two prompt expansion modes.
fal.live and Infiniteslop turn faster-than-playback clips into continuous interactive channels.
fal reports under three seconds of generation time for a five-second 768p clip with native audio.
The omni-modal base model ships with open weights, native stereo audio, and output up to 2K.