Find any model, tool or page and jump to it.
FLUX 3 is Black Forest Labs' state-of-the-art multimodal video model. One model draws the picture and writes the sound in the same pass. Write a prompt and generate a clip of five to twenty seconds at 720p or 1080p, with speech, effects, and ambience already in sync. Attach one image and it becomes the opening frame, pixel for pixel. Attach two and they become the first frame and the last. Attach up to ten and they become a storyboard the clip moves through in order. Attach a finished clip instead and FLUX 3 carries it on from its final frames, holding the momentum and the framing. Turn on draft mode to try ten directions fast, then render the one you want at full quality. Available in Visual Sandbox alongside other advanced AI video models, so you can generate, edit, and iterate in a single place. A natural fit for ad creative, music videos, storyboards, dialogue scenes, and social content.
| Tier | Details | Price |
|---|---|---|
| 720p | Text or image input | $0.22/s |
| 1080p | Text or image input | $0.37/s |
| 720p | Draft | $0.08/s |
| 720p | Continue a video | $0.52/s |
| 1080p | Continue a video | $0.67/s |
| 720p | Continue a video, draft | $0.15/s |
Billed per second of output. 1080p costs more than 720p, and continuing a video costs more than starting from a prompt or an image. Draft mode renders at 720p and costs about a third of the standard 720p rate. Visual Sandbox shows the exact price before you generate and refunds any unused portion when the clip finishes.