SD 2 First & Last Frame (Pro) by ByteDance. Generate video that transitions between two reference images. Provide 1 image for start-frame-only, or 2 images for both start and end frames.
SD 2 First & Last Frame (Pro) by ByteDance. Generate video that transitions between two reference images. Provide 1 image for start-frame-only, or 2 images for both start and end frames.
About this model
SD 2 First & Last Frame (Pro) generates video that transitions between reference images. Provide one image to use as the opening frame, or two images to constrain both the start and end of the video. Powered by ByteDance's SD 2 model with up to 2K resolution and 4–15 second duration.
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.25 per second (e.g. $1.25 for 5 seconds) | Pay per second of video generated. |
| Fal.ai | $0.3024/sec (high) / $0.2419/sec (basic) | Fal.ai charges $0.3024/sec for high quality and $0.2419/sec for basic. muapiapp is 17% cheaper on high ($0.25/sec) and 50% cheaper on basic ($0.12/sec equivalent). |
| Replicate | $0.3024/sec (high) / $0.2419/sec (basic) | Replicate charges the same as Fal.ai — $0.3024/sec (high), $0.2419/sec (basic). muapiapp saves you 17–50% depending on quality tier. |
Pay per second of video generated.
Fal.ai charges $0.3024/sec for high quality and $0.2419/sec for basic. muapiapp is 17% cheaper on high ($0.25/sec) and 50% cheaper on basic ($0.12/sec equivalent).
Replicate charges the same as Fal.ai — $0.3024/sec (high), $0.2419/sec (basic). muapiapp saves you 17–50% depending on quality tier.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt | string | Text description guiding the transition between frames. | Two people having a street interview, the interviewer holds a microphone. |
| Frame Images | array | 1 image = first frame only; 2 images = first and last frame. Use 'adaptive' aspect ratio to match the reference image geometry. | https://d3adwkbyhxyrtq.cloudfront.net/ai-images/186/712345784292/4a8c5c70-abcc-4920-873e-b0e219986453.jpg |
| Aspect Ratio | Enum (7 options) | Output video aspect ratio. 'adaptive' matches the reference image (recommended); concrete ratios may crop or pad. | adaptive |
| Duration (seconds) | int | Video duration in seconds. | 5 |
Text description guiding the transition between frames.
Two people having a street interview, the interviewer holds a microphone.1 image = first frame only; 2 images = first and last frame. Use 'adaptive' aspect ratio to match the reference image geometry.
https://d3adwkbyhxyrtq.cloudfront.net/ai-images/186/712345784292/4a8c5c70-abcc-4920-873e-b0e219986453.jpgOutput video aspect ratio. 'adaptive' matches the reference image (recommended); concrete ratios may crop or pad.
adaptiveVideo duration in seconds.
5Developer documentation
Upload 1 or 2 images: Provide images via images_list. The first image is the opening frame; the second (optional) is the closing frame.
Write your prompt: Describe the action or transition that should occur between the frames.
Set duration: Between 4 and 15 seconds.
Submit and poll: Use the returned request_id to poll for results. Aspect ratio is determined by the input images.
Frequently asked
No. Providing one image uses it as the first frame only. Providing two images constrains both the start and end frames.
No — the output aspect ratio is determined automatically by the dimensions of your reference images.
JPEG, PNG, WebP, and BMP are supported.