About FLUX 3 Max: Keyframes to Video
FLUX 3 Max (Keyframes to Video) generates a video with native audio that passes through up to 10 keyframe images, with the prompt describing the motion between them. FLUX 3 Max is a post-trained variant of FLUX 3 Video, built by fal with Black Forest Labs and tuned for stronger prompt adherence, better visual quality and much lower latency.
It generates 5 to 20 second clips at 720p or 1080p with synchronized audio, in aspect ratios from 21:9 to 9:16. A reasoning level controls how much the model expands the prompt before generating: high for closer prompt adherence, low for faster turnaround.
Ready to build with FLUX 3 Max: Keyframes to Video?
Try FLUX 3 Max: Keyframes to Video in the Workbench to prompt it, compare outputs, and iterate on prompts without writing any code. When you're ready to ship, call the same model from our API and build your own apps on top of it.
curl -sSf -X POST https://hub.oxen.ai/api/ai/videos/generate \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $OXEN_API_KEY" \
-d '{
"model": "flux-3-max-keyframes-to-video",
"prompt": "The woman picks up a brown felt fedora, lifts it onto her head with both hands and adjusts the brim, smiling at the camera. Soft studio light against a plain gray wall.",
"input_image": [
"https://hub.oxen.ai/api/repos/ox/Oxen-AI-Assets/file/main/images/ox_zoom_out_1926_1076.png"
],
"aspect_ratio": "auto",
"resolution": "720p",
"duration": 5,
"generate_audio": true,
"reasoning_level": "high"
}'Request parameters
Every field FLUX 3 Max: Keyframes to Video accepts in the request body. See the API reference for response formats.
https://hub.oxen.ai/api/ai/videos/generate| Field | Type | Required | Default | Description |
|---|---|---|---|---|
prompt | string | Yes | — | Describe how the video moves between the keyframes, including any sound. |
input_image | array<string> | Yes | — | Images the video passes through, 1 to 10 in order: the first starts the video, the last ends it, and the rest fall evenly in between. PNG, JPEG or WebP. With aspect ratio on auto, the video follows the first image's shape. |
aspect_ratio | string | No | "auto" | auto21:92:116:94:31:13:49:16 |
resolution | string | No | "720p" | 720p1080p |
duration | integer | No | 5 | Length of the generated video. Range: 5 – 20 |
generate_audio | boolean | No | true | Generate synchronized audio for the video. |
reasoning_level | string | No | "high" | How much effort the model spends expanding the prompt before generating. High follows the prompt more closely; low returns faster.lowhigh |
seed | integernullable | No | — | Random seed. A random seed is selected when omitted. |