Fal presents the FLUX 3 Max video model post-trained with Black Forest Labs – Unite.AI

fal announced FLUX 3 Max on October 19, 2026, a post-trained video generation model from fal Research in collaboration with Black Forest Labs on the FLUX 3 foundation model. The model became available the same day through fal’s inference platform, and fal reports that it combines top-tier image-to-video quality with much faster generation speeds than competing models with similar quality ratings.

How Fal Research and Black Forest Labs built the model

The launch pairs two research groups with distinct roles: Black Forest Labs, which developed the foundational FLUX 3 model, and fal Research, whose work spans post-training, reinforcement learning, inference optimization, and systems engineering. fal Research worked directly with the FLUX 3 model, continuing its training to improve the generation’s visual fidelity, motion consistency, grip and overall performance.

In the announcement, Black Forest Labs co-founder and CEO Robin Rombach said FLUX 3 was built as a foundation for others to build upon, and described fal’s customization of the model for developers and enterprises on the fal platform as the type of collaboration FLUX was designed for. Burkay Gur, CEO of fal, said the launch combines Black Forest Labs’ underlying model with his team’s research into inference quality and performance, adding that he believes some of the most important advances in generative AI will come from optimizing models and the systems that drive them together.

The Black Forest Labs home page features a banner stating that FLUX 3 Max is 22 times faster than Seedance 2.5 at the same quality and directs visitors to the fal product page.

Fal’s FLUX 3 Max product page describes three levels of work behind the model: post-training conducted with Black Forest Labs; inference optimization that changed how the model runs to reduce build latency and increase throughput while preserving quality; and systems engineering that optimized the service path on FAL’s inference engine for production workloads.

FLUX 3 Max follows H3 Max, the post-trained video model previously released by fal Research. fal said the H3 Max produced video in 3.49 seconds in its benchmark, which it described as generating faster than real time, while achieving top-tier image-to-video quality in independent evaluations. The product page notes that the H3 Max, at 7.55 seconds, is the only faster model in the published comparison.

Kaan Oktay, head of fal research, said the team achieved a 50-point Elo improvement while making inference more than 11 times faster, and that fal research is continuing to post-train the model. fal describes this launch as the first iteration of FLUX 3 Max, with further improvements to visual quality, build speed, and overall performance underway.

Quality and speed data reported

fal reported that FLUX 3 Max achieved an Elo rating of 1,139 out of 4,868 votes in independent preliminary assessments of human preferences on Artificial Analysis Video Arena, a result that the company says places the model among the highest quality image-to-video systems evaluated. Fal’s latency benchmark in the release measured generation of a 10-second video at 10.95 seconds, which the company described as 22 times faster than the Seedance 2.5’s 277.80 seconds with the same reported Elo rating, and 11.5 times faster than the original FLUX 3’s 126.44 seconds. The release also reports an image-video Elo gain of 50 points compared to the original FLUX 3, from 1,089 to 1,139.

The product page publishes a different latency value. It lists an average of 12.45 seconds for a 10-second 720p video, describing the improvement over FLUX 3 as about 10 times faster, and a comparison table listing Gemini Omni Flash 1.1 at 1,134 Elo and 47.37 seconds, MiniMax H3 at 768p at 1,141 Elo and 83.26 seconds, Wan 3.0 at 1,127 Elo and 180.16 seconds, Seedance 2.5 at 1,139 Elo and 277.80 seconds and the original FLUX 3 at 1,089 Elo and 126.44 seconds. The page identifies the Elo data as preliminary AA-Video-I2V v1.1 ratings from Artificial Analysis and the build times as averages for a 10-second 16:9 video, and describes the FLUX 3 Max as ranking first in image-to-video quality, while noting that the MiniMax H3 is listed first on that listing.

Artificial Analysis’ Image-Video Ranking, which ranks models by Elo derived from blind-paired human preference votes cast by validated raters, listed the 720p FLUX 3 Max with an Elo rating of 1,138 plus or minus 15 out of 5,054 samples, in second place behind the 768p MiniMax H3 with 1,140 Elo. The board lists Seedance 2.5 at 1,138 Elo and the original FLUX 3 at 1,086 Elo, and dates the release of FLUX 3 Max to October 2026.

Availability and launch prices

FLUX 3 Max is served via fal’s inference platform under a unified API. The ad describes access to text-to-video, image-to-video, and keyframe-to-video generation, while the product page lists two endpoints at launch, text-to-video and image-to-video, both at 720p. According to the page, content generated via the fal.ai API can be used in commercial projects under fal’s terms.

The launch price is set at $0.0425 per second of video output at 720p, billed per second with no minimums or subscription. The product page states that the rate is reduced by 75% from the list price of $0.17 per second and lasts until November 7, 2026, and fal’s release describes the discount as valid for the first three weeks.