SD 2.0 Extend Video continues an existing SD 2.0 generated video seamlessly. Provide the original request ID and an optional prompt to guide the extension — the model preserves visual style, motion, characters, and audio consistency across the new segment. Optional image, video, and audio references can be supplied to steer the extension: user-supplied references map to @image2…@image9, @video1…@video3, @audio1…@audio3 in the prompt (the source video's last frame is always @image1).
About this model
SD 2.0 Extend Video seamlessly continues an existing SD 2.0 generated video. Provide the original request ID and an optional prompt to guide the new segment — the model preserves visual style, motion physics, character identity, and native audio across the extension. Ideal for building longer narratives, adding follow-up scenes, or exploring alternate continuations of a generated clip.
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.60 per video | muapiapp offers SD 2.0 Extend starting at $0.60 per video (5s, basic quality), scaling at $0.12/sec for basic and $0.25/sec for high quality across 5–15 second durations. |
| 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). |
| 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. |
muapiapp offers SD 2.0 Extend starting at $0.60 per video (5s, basic quality), scaling at $0.12/sec for basic and $0.25/sec for high quality across 5–15 second durations.
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).
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 |
|---|---|---|---|
| Request Id | string | Request ID of the original Seedance 2.0 video generation. | cab9517f-1818-4910-8d66-292701c78c2d |
| Prompt | string | Optional prompt to guide the extension. Reference additional images with @image2…@image9, videos with @video1…@video3, and audio with @audio1…@audio3 — the source video's last frame is always @image1. | |
| Image URLs | array | Up to 8 additional reference image URLs (JPEG/PNG/WebP). Each Nth image corresponds to @image(N+1) in the prompt (the source video's last frame is @image1). | undefined |
| Video Reference URLs | array | Up to 3 reference video clip URLs (MP4, max 15s each). Each Nth video corresponds to @videoN in the prompt. | undefined |
| Audio Reference URLs | array | Up to 3 reference audio clip URLs (MP3/WAV, total max 15s). Each Nth audio corresponds to @audioN in the prompt. | undefined |
| Aspect Ratio | Enum (6 options) | Output video aspect ratio (only used when reference images/videos/audio are provided). | 16:9 |
| Duration (seconds) | int | Length of the extension clip in seconds. | 5 |
| Quality | Enum (2 options) | - | basic |
Request ID of the original Seedance 2.0 video generation.
cab9517f-1818-4910-8d66-292701c78c2dOptional prompt to guide the extension. Reference additional images with @image2…@image9, videos with @video1…@video3, and audio with @audio1…@audio3 — the source video's last frame is always @image1.
Up to 8 additional reference image URLs (JPEG/PNG/WebP). Each Nth image corresponds to @image(N+1) in the prompt (the source video's last frame is @image1).
undefinedUp to 3 reference video clip URLs (MP4, max 15s each). Each Nth video corresponds to @videoN in the prompt.
undefinedUp to 3 reference audio clip URLs (MP3/WAV, total max 15s). Each Nth audio corresponds to @audioN in the prompt.
undefinedOutput video aspect ratio (only used when reference images/videos/audio are provided).
16:9Length of the extension clip in seconds.
5-
basicDeveloper documentation
Get the Request ID: Generate a video using sd-v2.0-t2v or sd-v2.0-i2v and copy the request_id from the response.
Optionally Add a Prompt: Describe what should happen next in the video. If left empty, the model intelligently continues the existing scene.
Choose Quality: Select basic ($0.12/sec) for drafts or high ($0.25/sec) for cinema-grade output.
Set Duration: Choose 5, 10, or 15 seconds for the extended segment.
Submit and Poll: You'll receive a new request_id. Poll the result endpoint until status is completed.
Frequently asked
You can extend any video generated by SD 2.0 (sd-v2.0-t2v or sd-v2.0-i2v). Provide the request_id returned by the original generation.
No, the prompt is optional. If omitted, SD 2.0 intelligently continues the scene based on the original video's content, motion, and style.
Yes. The model preserves visual style, character identity, motion physics, camera movement, and native audio consistency across the extension.
SD 2.0 supports up to 2K resolution output, maintained through the extension.