As of: 2026-10-25 23:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/m/flux-3-max/

# FLUX 3 Max

Black Forest Labs, FLUX 3 family. Status: current. Type: video generation. Released: 2026-10-19.

## How to call it

| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| fal | `blackforestlabs/flux-3-max/text-to-video` |  | https://fal.ai/models/blackforestlabs/flux-3-max/text-to-video |
| fal | `blackforestlabs/flux-3-max/image-to-video` |  | https://fal.ai/models/blackforestlabs/flux-3-max/image-to-video |
| fal | `blackforestlabs/flux-3-max/keyframes-to-video` |  | https://fal.ai/models/blackforestlabs/flux-3-max/keyframes-to-video |

## Pricing

- Per second 720p: $0.0425
- Per second 1080p: $0.0725
- Unit as published: USD per second of output video on fal, 75% launch discount (Oct 15 - Nov 7, 2026); list price after Nov 7: 0.17/s 720p, 0.29/s 1080p. Text-to-video and image-to-video same rate
- Source: https://fal.ai/models/blackforestlabs/flux-3-max/text-to-video
- Last checked: 2026-10-25

## Specs

- Input: text, image
- Output: video, audio
- Context window: not published
- Max output: not published
- Open weights: no
- Licence: proprietary

## Where it stands

- No newer video generation model in the registry yet.
- Before it from Black Forest Labs: FLUX 3 (2026-07-23) https://postcutoff.com/m/flux-3/
- Closed weights; access is hosted only.

## What stands out

- Fast frontier-quality video: fal says a 10 s 720p clip takes about 12.45 s, versus 126.44 s for FLUX 3 and 277.80 s for Seedance 2.5 (fal's own measurement). Source: https://fal.ai/flux-3-max
- Post-trained FLUX 3 Video by fal Research: Not a separate BFL tier: fal Research post-trained BFL's FLUX 3 Video with BFL for prompt adherence and lower latency; +50 Elo over FLUX 3 on AA image-to-video per fal. Source: https://x.com/fal/status/2112251022047412492
- Keyframes-to-video: Up to ten images placed at chosen frame indexes, with a prompt describing the motion between them; 5-20 s, 720p or 1080p, audio on by default. Source: https://fal.ai/models/blackforestlabs/flux-3-max/keyframes-to-video

## Notes

Artificial Analysis text-to-video leaderboard (read 2026-10-25): #6, Elo 1155, listed as 'FLUX 3 Max (720p)' via fal; AA price 2.55 USD per minute (= fal promo 0.0425/s at 720p). Distinct from flux-3 (BFL's own FLUX 3 Video, AA t2v #8, Elo 1145, BFL API flux-3-video). Announced by fal on X 2026-10-19 (444k views by 2026-10-25); fal's pricing page dates the promo from Oct 15. Not seen on BFL's own API or pricing page, nor on OpenRouter; served only by fal as far as checked. No timeline entry.

fal's post-trained, low-latency edition of Black Forest Labs' FLUX 3 Video, with native audio. Call it through fal's queue API, e.g. `fal.subscribe("blackforestlabs/flux-3-max/text-to-video", { input: { prompt, duration, aspect_ratio, generate_audio } })`.

Artificial Analysis text-to-video (with audio) leaderboard, read 2026-10-25: rank 6, Elo 1155 (https://artificialanalysis.ai/video/leaderboard/text-to-video). See also the base model, FLUX 3 (model id flux-3).

Sources: https://fal.ai/flux-3-max · https://fal.ai/models/blackforestlabs/flux-3-max/text-to-video · https://fal.ai/models/blackforestlabs/flux-3-max/keyframes-to-video · https://x.com/fal/status/2112251022047412492

Model page: https://postcutoff.com/m/flux-3-max/