Introduction
Send a prompt on its own to get a 5-second 720p clip in 16:9 with sound, or add duration_seconds, resolution, aspect_ratio, generate_audio, watermark, return_last_frame and output_format to steer it. Nothing needs to be uploaded.
| Field | Value |
|---|---|
| Input | Text |
| Output | Video |
| Publisher | ByteDance (Seedance 2.5 Flash, speed tier of Seedance 2.5, announced by BytePlus on 12 November 2026) |
| API names | bytedance/seedance-2.5-flash/text-to-video, image-to-video, reference-to-video, video-edit, video-extend |
| Resolution | 480p or 720p (default 720p); 1080p and 4K not supported |
| Clip length | 4–30 whole seconds, default 5; video-edit follows its 4–30 s source |
| Aspect ratio | 16:9 (default), 9:16, 4:3, 3:4, 1:1, 21:9 or adaptive on text and reference requests; image-to-video follows the start image |
| Reference limits | Up to 10 images, 5 clips (30 s in total) and 5 audio files on reference-to-video |
| Reference clip rules | Each clip counts as at least 2 s and needs about 409,600 pixels (854×480 or larger) |
| Edit inputs |
|
| Audio |
|
| Output options | mp4 (default) or mov; watermark off by default; |
| Billing basis | Per second by resolution; reference clips add their seconds; edits bill source seconds; failed generations are free |
| Measured render time | 84 s to about 11 min for 4 s at 480p (13 November 2026 runs) |
Key features & highlights
Flagship workflows at a lower ceiling
Flash keeps the four Seedance 2.5 workflows that fit drafting (text, start and end frames, mixed references, clip editing), drops 1080p and 4K, and adds a video-extend built on its image-to-video mode. BytePlus presents the trade as results close to the flagship at nearly twice the speed. Our 4-second 480p renders took anywhere from 84 seconds to about 11 minutes, so measure speed on your own queue.
Give every upload a job in the prompt
Name uploads by type and order with a space: @Image 1, @Video 1, @Audio 1. With two portraits as @Image 1 and @Image 2 in 16:9, both people appeared together in one elevator scene and kept their faces, hair and clothes. With one image, one clip and one voice file, the clip's street, yellow tram and tracking camera carried over, but the person stayed back to the camera and the face of @Image 1 barely showed.
A sunny street turned into a snowy evening
Both riders, their bikes, the tram and the camera move matched the source frame by frame, and the weather and light changed while nothing else did. That was a 4.0 s clip at 480p with audio off, which also kept the original soundtrack.
One switch decides whose soundtrack you get
generate_audio is on by default and produces speech, effects and ambience with the picture. In video-edit, switching it off keeps the source clip's own track (44.1 kHz in our test), while switching it on generates a new one that can follow @Audio 1. The switch has no effect on the price.
Request parameters
| Parameter | Required | Description |
|---|---|---|
prompt | Yes | Write the subject, one main action, the camera move, the light and a closing Audio: line that lists what is heard; the tested prompt followed that order and the clip matched it. at least 1 characters · at most 20000 characters |
aspect_ratio | No | 16:9 (default), 9:16, 4:3, 3:4, 1:1, 21:9 or adaptive, which leaves the frame shape to the model. |
resolution | No | 480p or 720p, default 720p. A widescreen request at 480p returned 854×480, so settle the shot at 480p and repeat the winning prompt at 720p. |
duration_seconds | No | Whole seconds from 4 to 30, default 5, billed as requested. If you plan to cut or chain the clip, ask for a second more than you need; our 4-second request returned 4.064 seconds. |
generate_audio | No | True by default: speech, effects and ambience from the prompt are produced with the picture, and our test returned an AAC stereo track. Default: |
output_format | No | mp4 (default) or mov. BytePlus documents mov as H.264 video with yuv444p chroma sampling and PCM audio, which suits an editing timeline; our test clips were mp4. |
return_last_frame | No | False by default. Default: |
watermark | No | False by default. BytePlus, ByteDance's cloud platform, says true adds an AI-generated mark to the lower-right corner of the video, so leave it off for files you will edit. Default: |
Pricing
Billed per requested second at the 480p or 720p rate; a 4-second request that returns 4.06 seconds is billed as 4. Audio, aspect ratio, watermark, last frame and file format do not change the price, and failed requests are not charged.
| Specification | Price |
|---|---|
| $0.115 / second |
| $0.2392 / second |
Use cases
Trying a shot plan before a final render
Send a 4-second, 480p, 16:9 prompt that ends with an Audio: line, such as a rain-soaked night crosswalk. Our run returned 854×480 at 24 fps, 4.064 s long, with an AAC stereo track of rain and laughter. Change one phrase and rerun to see what moved.
Restyling footage you already shot
Upload a 4–30 second clip, describe the change, then list what must stay. A sunny street with two cyclists and a tram became a snowy evening in our test, and with generate_audio switched off the original soundtrack came back unchanged.
Character and product mockups
Attach a portrait and a product photo as @Image 1 and @Image 2 and ask for 9:16. Our two-image test returned 480×854 with the man lifting a glass dropper bottle toward the lens, so both references showed up. Use the result to judge the idea before building a polished version.
Looping a portrait animation
Give the same photo as image_url and last_image_url and describe the motion between them. Our potter portrait returned a 562×748 clip of 4.064 s whose opening and closing frames both matched the source pose.
Chaining one shot into the next
Set return_last_frame to true and the MP4 arrives with a JPEG of its closing frame (854×480 at 480p 16:9 in our test). Pass that JPEG as image_url of the next image-to-video request so the following shot begins where the last one stopped.
Prompt guide & pro tips
Choose the endpoint for your input
Use text-to-video for a prompt alone, image-to-video for a start frame, reference-to-video for images, clips or a voice, video-edit for an existing 4–30 second clip, and video-extend to add 4–30 seconds to a clip of up to 60 seconds. All five share the 480p/720p range and the
generate_audioswitch.Send the createTask request
POST to https://api.spicyapi.ai/api/v1/jobs/createTask with the header Authorization: Bearer YOUR_SPICY_API_KEY and the body {"model": "bytedance/seedance-2.5-flash/text-to-video", "input": {"prompt": "A lighthouse keeper climbs a spiral stair at dawn and wipes fog from the lamp glass. Audio: wind, boots on iron.", "
duration_seconds": 4, "resolution": "480p"}}.Poll for the finished clip
The response gives a taskId in state queued. Call GET /api/v1/jobs/recordInfo?taskId=YOUR_TASK_ID, or pass a callBackUrl, until the state is succeeded, then read the MP4 address from output.assets. Field details are in the task guide at docs.spicyapi.ai/docs/tasks.
Change one phrase, then move up
Compare variations at 480p and 4 seconds, altering a single phrase per run. When a take works, rerun it at 720p and your target length, or send the same brief to Seedance 2.5 for 1080p.
Seedance 2.5 Flash text-to-video from a written shot
Seedance 2.5 Flash text-to-video is ByteDance's speed tier of Seedance 2.5 for prompt-only generation: send text, receive a clip of 4 to 30 seconds at 480p or 720p, with speech, effects and ambience generated in the same pass. BytePlus, ByteDance's cloud platform, says Flash gives results close to the flagship at nearly twice the generation speed, and marks 1080p and 4K as not supported for it.
Request example
Create a task with the public name bytedance/seedance-2.5-flash/text-to-video. Send POST https://api.spicyapi.ai/api/v1/jobs/createTask with Authorization: Bearer YOUR_SPICY_API_KEY and Content-Type: application/json; the task API guide covers polling and callbacks.
{
"model": "bytedance/seedance-2.5-flash/text-to-video",
"input": {
"prompt": "A young woman in a yellow raincoat sprints across a rain-soaked city crosswalk at night, neon signs reflected in the wet asphalt. She ducks under a shop awning, pushes back her hood and laughs, shaking water from her hair. Handheld tracking shot at street level. Audio: heavy rain on pavement, a passing car hissing through puddles, her breathless laugh.",
"duration_seconds": 4,
"resolution": "480p",
"aspect_ratio": "16:9",
"generate_audio": true,
"return_last_frame": true
}
}
Response example
Submission returns a queued task; poll GET /api/v1/jobs/recordInfo?taskId=YOUR_TASK_ID (or pass callBackUrl) until it reads succeeded.
{"code": 200, "msg": "success", "data": {"taskId": "YOUR_TASK_ID", "state": "queued"}}
{"code": 200, "msg": "success", "data": {"taskId": "YOUR_TASK_ID", "state": "succeeded", "output": {"assets": [{"url": "https://example.com/clip.mp4"}]}}}
With return_last_frame on, the finished task also returns a JPEG of the closing frame.
Parameters
| Parameter | Values | Default |
|---|---|---|
prompt | Text brief | Required |
duration_seconds | 4 to 30 | 5 |
resolution | 480p, 720p | 720p |
aspect_ratio | 16:9, 9:16, 4:3, 3:4, 1:1, 21:9, adaptive | 16:9 |
generate_audio | true, false | true |
watermark | true, false | false |
return_last_frame | true, false | false |
output_format | mp4, mov | mp4 |
How billing works
Text-to-video is billed per requested second at the rate of the chosen resolution, and the page shows the price for one second. A 4-second request came back at 4.064 seconds and was billed as 4. Audio, ratio, watermark, last frame and format do not change it; a failed generation is not charged.
Writing a shot from text alone
Prompt order. Lead with the subject and its one main move, then camera, light and sound. BytePlus lists subject, action, scene, visual style, camera and sound and says unneeded parts can be dropped; Dreamina says to keep Flash prompts concise and focused on one clear action. The tested prompt stays on one street: a yellow raincoat, a run for cover, a handheld tracking shot, neon on wet asphalt.
Sound. Close with an Audio: line that lists sources in plain words. For speech, put the words in quotes and name the language and tone before them, the rule BytePlus gives for non-Chinese dialogue. BytePlus also documents brackets for music, effects and dialogue; we tested the plain line only.
Frame shape. Use 16:9 for landscape players, 9:16 for vertical feeds, 1:1, 4:3 or 3:4 for grids and 21:9 for wide crops. adaptive lets the model decide, which suits early exploration; set a fixed ratio when a template expects one.
Length. Our 4-second test held three beats: run, take cover, laugh. For 8 to 15 seconds plan two or three beats under one camera idea. For 16 to 30 seconds write time-ranged lines (0-8 s: ..., 8-20 s: ...); ByteDance Seed says prompts can steer narrative, camera and rhythm for a specific time frame. We ran 4-second requests alone, so test longer plans at 480p before moving to 720p.
Chaining. Set return_last_frame to true, then pass the returned JPEG as image_url to bytedance/seedance-2.5-flash/image-to-video with a prompt for the next beat; the start image sets the shape, so a 16:9 frame continues in 16:9. Each clip gets its own soundtrack, so reuse your Audio: wording. Many generators ease motion near a clip's end, so request a second more than you need and trim.
File settings. Use mp4 for playback and mov for an editing timeline; BytePlus documents mov as H.264 video with yuv444p chroma sampling and PCM audio. Keep watermark off unless a platform requires a mark, since BytePlus says it adds an AI-generated mark to the lower-right corner.
What our tests showed
| Run (13 Nov 2026) | Result |
|---|---|
| 4 s, 480p, 16:9, audio on | 854×480, 24 fps, 4.064 s, AAC stereo; she ran across, ducked under the awning and laughed, as written. Done in 2.6 minutes on one run, about 11 minutes on another. |
Same plus return_last_frame | The MP4 and an 854×480 JPEG of the final frame. |
Across all four Flash endpoints, 4-second 480p runs took 84 seconds to about 11 minutes, so we cannot confirm a speed gain; longer and 720p clips take longer. We did not run a full multi-shot chain or test mov. A duplicate submission starts a second render, so wait on the first task.
Limits and errors
- Values outside 4 to 30 seconds,
480por720p, or the listed ratios are refused at submission. - The model runs its own copyright and likeness check, and one of our text requests was stopped by it. Rewrite with an invented character instead of a named person or protected character; a failed task carries an error message and is not charged.
- ByteDance Seed says complex motion physics and scenes with several interacting subjects still have room to improve, so keep a shot to one or two subjects.
- Flash has no 1080p, 4K or video-extend endpoint. Finished files stay available for 14 days.
When to pick another endpoint or model
- Seedance 2.5: delivery above 720p, 4K, clip extension, or draft-then-upscale.
- Seedance 2.0 Fast: the previous-generation fast tier at 480p and 720p, with an extend endpoint Flash lacks.
- Flash image-to-video: a picture must open the shot, or you continue from a returned last frame.
- Flash reference-to-video: faces, products, camera moves or voices come from your own files.
- Flash video-edit: restyle footage you already have.
API quick start
# Generate the idempotency key once; reuse it when retrying after a lost response.
export SPICY_IDEMPOTENCY_KEY="${SPICY_IDEMPOTENCY_KEY:-$(uuidgen)}"
curl -X POST https://api.spicyapi.ai/api/v1/jobs/createTask \
-H "Authorization: Bearer $SPICY_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $SPICY_IDEMPOTENCY_KEY" \
-d '{
"model": "bytedance/seedance-2.5-flash/text-to-video",
"input": {
"prompt": "A slow aerial shot drifting over a misty pine forest at sunrise",
"resolution": "720p",
"aspect_ratio": "16:9",
"output_format": "mp4",
"duration_seconds": 5
}
}'
# Copy data.taskId from the response into TASK_ID, then poll the status.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.spicyapi.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
-H "Authorization: Bearer $SPICY_API_KEY"
See the API reference for authentication, the complete schema, polling, and error handling.










