FLUX.3 Text to Video

FLUX.3 Text to Video is a text to video model. A written description becomes footage. No reference needed. People run it for: frontier, faceless, native audio, long clip.

10 credits for a 5 s clip. It bills per second of output at 2 credits a second, so a shorter clip costs proportionally less. Billing stops at 20 seconds, so one run never costs more than 40 credits. Credits are the only unit we quote. One balance covers every model in the catalogue, so there is no subscription attached to FLUX.3 Text to Video specifically and no minimum spend to use it. See the pricing page for what a credit costs.

It accepts a text prompt. It will not start without a text prompt. Resolutions: 720p, 1080p. Aspect ratios: 21:9, 2:1, 16:9, 4:3, 1:1, 3:4, 9:16.

Every run returns one video. A single run gives you between 5 and 20 seconds. It produces the audio along with the picture. Runs are queued and processed in the background, so you can close the tab — the finished file turns up in your library. We do not quote a fixed wall-clock time: it moves with the length of the run and with how busy the queue is.

The catalogue also runs HappyHorse 1.1 Text to Video (10 credits for a 5 s clip), Kling 3.0 Pro Text to Video (10 credits for a 5 s clip) and Wan 2.7 Text to Video (10 credits for a 5 s clip) for the same job. Kling 3.0 Pro Text to Video is our default pick for text to video. They all draw on one credit balance, so trying a second one costs nothing beyond the run itself.