Launch sources · checked October 19, 2026
Use the FLUX 3 Max API
Use fal’s 3 FLUX 3 Max endpoints for text-to-video, image-to-video or keyframes-to-video. Image-to-video requires prompt and image_url; it supports 5–20 seconds at 720p or 1080p with native audio enabled by default.
Endpoints
blackforestlabs/flux-3-max/text-to-videoblackforestlabs/flux-3-max/image-to-videoblackforestlabs/flux-3-max/keyframes-to-video
Choose an endpoint, prepare the required inputs, then follow fal’s official API documentation to submit your request. This site has no FLUX 3 Max API integration.
Image-to-video parameters
| Parameter | Value |
|---|---|
| prompt | Required text prompt |
| image_url | Required image URL: PNG, JPEG or WebP |
| duration | Integer 5–20 seconds; default 5 |
| resolution | 720p (default) / 1080p |
| aspect_ratio | auto (default) · 21:9 · 2:1 · 16:9 · 4:3 · 1:1 · 3:4 · 9:16 |
| generate_audio | true (default) |
| reasoning_level | low / high (default: high) |
| end_image_url | Optional last frame URL |
| seed | Optional seed |
Questions and answers
Can I supply a last frame?
Yes. The image-to-video endpoint accepts optional end_image_url alongside the required image_url and prompt.
Is audio enabled by default?
Yes. generate_audio defaults to true in the fal image-to-video schema.
Can I use FLUX 3 Max through kie.ai today?
No. As checked October 19, 2026, kie.ai does not offer FLUX 3 Max. Use fal for this model.