FLUX 3 Max is fal’s post-trained text-to-video and image-to-video model, built with Black Forest Labs on the FLUX 3 base. fal says it generates a 10-second 720p clip in about 12 seconds, around 10x faster than FLUX 3. It is tied for #1 on a preliminary Artificial Analysis image-to-video board. Pricing is $0.0425 per second of video through November 7 (list $0.17), with a playground and API.
H3 Max is fal's post-trained MiniMax H3, ranked #1 for quality, prompt understanding, and aesthetics against 12 leading video models, while generating a 5s video in ~3 seconds (35x the throughput of official H3).
Post-trained with new data on prompt adherence and visual quality, co-optimized with fal's inference stack on NVIDIA GB200 NVL72. Beats Gemini Omni Flash, Wan 3.0, Kling 3, and Veo 3.1 head-to-head.
