Create video
AvailableText-to-video and image-to-video, with first and last frames and reference images. From 2 credits per second, depending on the model, resolution and sound.
What it does
Generate a video from a prompt, animate an image (image_url is the first frame, end_image_url the last), or keep subjects consistent with reference_image_urls. Models: gemini-omni-flash (default): video with sound; veo-3.1: Veo at up to 4K; veo-3.1-fast: Veo quality at a lower price, up to 4K; veo-3.1-lite: cheap clips with sound, or silent b-roll; wan-3: single shots up to 30 s at native 1080p; happyhorse-1.1: talking characters with lip-sync; minimax-h3-max: fast image-to-video with sound; minimax-h3-max-turbo: the cheapest H3 Max clips; minimax-h3: the open-weights H3, up to 4K; seedance-2.5: ads up to 30 s with many reference images; seedance-2: cinematic multi-shot clips; seedance-2-fast: Seedance 2.0 at a lower price; seedance-2-mini: cheap Seedance drafts; kling-3-pro: cinematic motion; kling-3-turbo: faster Kling with sound; kling-o3: consistent characters from reference images; grok-imagine-video-1.5: short clips with sound from text or images; flux-3: cinematic clips from text or a first and last frame; ltx-2.5: open-weights clips up to 20 s and 4K; pixverse-6: cheap social clips with optional sound; vidu-q3: clips up to 16 s with start and end frames; luma-ray-3.2: cinematic silent shots; creatify-boreal: product, UGC and presenter ads from a script. Duration, resolution and aspect ratio are fitted to what the model makes, and any change is listed in warnings. Returns a job_id at once: videos take about 1–5 minutes, so call get_job with that job_id to get the permanent URL. Call list_models for prices and describe_model for each model's options. Quotes above 500 credits need confirm: true.
Parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
| prompt | string | Required | — | What happens in the video: subject, motion, camera, soundUp to 5,000 characters |
| model | gemini-omni-flash | veo-3.1 | veo-3.1-fast | veo-3.1-lite | wan-3 | happyhorse-1.1 | minimax-h3-max | minimax-h3-max-turbo | minimax-h3 | seedance-2.5 | seedance-2 | seedance-2-fast | seedance-2-mini | kling-3-pro | kling-3-turbo | kling-o3 | grok-imagine-video-1.5 | flux-3 | ltx-2.5 | pixverse-6 | vidu-q3 | luma-ray-3.2 | creatify-boreal | Optional | gemini-omni-flash | gemini-omni-flash (default): video with sound; veo-3.1: Veo at up to 4K; veo-3.1-fast: Veo quality at a lower price, up to 4K; veo-3.1-lite: cheap clips with sound, or silent b-roll; wan-3: single shots up to 30 s at native 1080p; happyhorse-1.1: talking characters with lip-sync; minimax-h3-max: fast image-to-video with sound; minimax-h3-max-turbo: the cheapest H3 Max clips; minimax-h3: the open-weights H3, up to 4K; seedance-2.5: ads up to 30 s with many reference images; seedance-2: cinematic multi-shot clips; seedance-2-fast: Seedance 2.0 at a lower price; seedance-2-mini: cheap Seedance drafts; kling-3-pro: cinematic motion; kling-3-turbo: faster Kling with sound; kling-o3: consistent characters from reference images; grok-imagine-video-1.5: short clips with sound from text or images; flux-3: cinematic clips from text or a first and last frame; ltx-2.5: open-weights clips up to 20 s and 4K; pixverse-6: cheap social clips with optional sound; vidu-q3: clips up to 16 s with start and end frames; luma-ray-3.2: cinematic silent shots; creatify-boreal: product, UGC and presenter ads from a script. Call list_models for what each model costs. |
| image_url | string | Optional | — | First frame: animate this image |
| end_image_url | string | Optional | — | Last frame: the video ends on this image (needs image_url on most models) |
| reference_image_urls | string[] | Optional | — | Images of subjects or products to keep consistent (not frames); see describe_model for each model's limit |
| duration | integer | Optional | — | Seconds (default 5); fitted to what the model makes1–30 |
| resolution | string | Optional | — | 360p, 480p, 540p, 720p, 768p, 1080p, 1440p, 2160p, 2k, 4k; fitted to the model's nearest; default: the model's standard one |
| aspect_ratio | string | Optional | — | "16:9", "9:16", "1:1"…; fitted to the nearest ratio the model makes |
| audio | boolean | Optional | — | Sound on (the default wherever the model can) or off; off costs less on some models |
| seed | integer | Optional | — | 0–2,147,483,647 |
| extras | object | Optional | — | Model-specific parameters passed through (see describe_model). Billed parameters (duration, resolution, aspect_ratio, audio, frames, references) are not accepted here; use the top-level fields. |
| confirm | boolean | Optional | false | Set true to accept a quote above 500 credits |
Which model should I use?
| Model | Best for | Price | |
|---|---|---|---|
| Gemini Omni Flash 1.1 | video with sound | 6 credits per second (360p) | |
| Veo 3.1 | Veo at up to 4K | 40 credits per second (720p silent) | |
| Veo 3.1 Fast | Veo quality at a lower price, up to 4K | 20 credits per second (720p silent) | |
| Veo 3.1 Lite | cheap clips with sound, or silent b-roll | 6 credits per second (720p silent) | |
| Wan 3.0 | single shots up to 30 s at native 1080p | 10 credits per second (480p) | |
| HappyHorse 1.1 | talking characters with lip-sync | 28 credits per second (720p) | |
| MiniMax H3 Max | fast image-to-video with sound | 10 credits per second (480p) | |
| MiniMax H3 Max Turbo | the cheapest H3 Max clips | 5 credits per second (480p) | |
| MiniMax H3 | the open-weights H3, up to 4K | 10 credits per second (480p) | |
| Seedance 2.5 | ads up to 30 s with many reference images | about 44.14 credits per second (480p, 880×500) | |
| Seedance 2.0 | cinematic multi-shot clips | about 28.88 credits per second (480p, 880×500) | |
| Seedance 2.0 Fast | Seedance 2.0 at a lower price | about 23.1 credits per second (480p, 880×500) | |
| Seedance 2.0 Mini | cheap Seedance drafts | about 14.44 credits per second (480p, 880×500) | |
| Kling 3.0 Pro | cinematic motion | 22.4 credits per second (silent) | |
| Kling 3.0 Turbo | faster Kling with sound | 28 credits per second | |
| Kling O3 | consistent characters from reference images | 22.4 credits per second (silent) | |
| Grok Imagine Video 1.5 | short clips with sound from text or images | 16 credits per second (480p) | |
| FLUX 3 | cinematic clips from text or a first and last frame | 12 credits per second (720p draft) | |
| LTX-2.5 | open-weights clips up to 20 s and 4K | 18 credits per second (720p) | |
| PixVerse V6 | cheap social clips with optional sound | 12 credits per second (720p with audio) | |
| Vidu Q3 | clips up to 16 s with start and end frames | 14 credits per second (360p) | |
| Luma Ray 3.2 | cinematic silent shots | 100 credits per clip (5 s 540p) | |
| Creatify Boreal | product, UGC and presenter ads from a script | 2 credits per second (720p) |
Examples
- A golden retriever running along a beach at sunset, waves crashing, 6 seconds with sound
- Animate this product photo: slow zoom in, soft studio lighting, no sound
- A video that starts on this photo and ends on that one, smooth camera motion between them
Use cases
- Turning a single product photo into a short video ad without a video shoot
- Prototyping motion and camera moves for a concept before a full production
- Interpolating between a first and last frame for a smooth transition shot
How to use it
Connect Sakaira
One click or one URL. Sign in with Google. No API keys.
Ask in plain language
"A golden retriever running along a beach at sunset, waves crashing, 6 seconds with sound"
Check the cost and balance in the reply
Every result tells you what it cost and what you have left.
FAQ
Which model should I use?
gemini-omni-flash is the default, and its videos always come with sound. Call list_models to see every video model available to you, from fast drafts to Veo 3.1 at up to 4K.
How much does it cost?
From 2 credits per second, depending on the model, resolution and sound. The reply always states the exact cost and your remaining balance.
Why do I get a job_id instead of the video right away?
Video takes longer to generate than an image, so create_video queues the job and replies with a job_id at once; call get_job with that id, or just ask "is my video done yet?", to get the result.
Ready to try create video?
Get 100 free credits