Cookie Preferences
We use cookies to personalize content and analyze traffic. See our Cookie Policy.
No video yet. Submit the form to generate video.
Explore different use cases and parameter configurations
Transparent pricing with no hidden fees. Pay as you go.
Billing for Video Models
You might notice that models "with video input" have a lower unit price. This is because the billing method differs:
| Rule & Modality | Channel | Credits | Price (USD) | Official / Reference Price | Daily Savings |
|---|---|---|---|---|---|
seedance-2-without-video-480p videoByteDanceresolution: 480p, reference_video_urls: empty | Starter | 10per second | $0.045 | ByteDance$0.151per second | Based on 1,000 10-second videos/dayByteDance costs $1055.450/day moreApiPass is 69.90% lower |
seedance-2-with-video-480p videoByteDanceresolution: 480p, reference_video_urls: not empty | Starter | 5.45per second | $0.025 | ByteDance$0.091per second | Based on 1,000 10-second videos/dayByteDance costs $662.270/day moreApiPass is 72.78% lower |
Competitor pricing costs $656.360/day more
ApiPass is 21.73% lower, estimated at Based on 1,000 10-second videos/day
seedance-2-without-video-720p
resolution: 720p, reference_video_urls: empty
ApiPass Price
$0.236
52 credits per second
Competitor Prices
Complete guide to using Seedance 2.0
Turn text, images, video clips, and audio into cinematic, audio-synced videos with a single API call.
Seedance 2.0 is ByteDance Seed's next-generation video model built on a unified multimodal audio-video joint generation architecture that accepts text, image, audio, and video inputs in a single call. Powered by a Dual-Branch Diffusion Transformer, the Seedance 2.0 API generates 4–15 second clips up to 1080p across aspect ratios from 16:9 to 9:16 — with sound effects, dialogue, music, and ambient atmosphere produced natively alongside the visuals, no post-production audio alignment required. Whether you're powering an AI ad generator, an avatar platform, or a creative studio backend, it delivers production-ready, reference-driven video generation out of the box.
Every request requires input.prompt. Text-only requests can omit all image, video, and audio reference fields.
Use first_frame_url and optionally last_frame_url to guide the video start and end frames. These fields cannot be combined with multimodal reference inputs.
Use reference_image_urls, reference_video_urls, and reference_audio_urls as style or content references. The adapter enforces the documented maximum item counts.
Combine up to 9 images, 3 video clips, and 3 audio files in a single generation. Reference them in your prompt as [Image1], [Video1], [Audio1], etc. Text, image, video, and audio all live inside one unified request.
Seedance 2.0 delivers significantly improved base capabilities with more realistic physics, natural motion, and precise prompt following. The model handles complex actions and continuous movements with stability, producing smoother, more authentic video output. The model executes what you describe instead of improvising off-script.
Supports text_to_video, first_last_frames (1-2 images as start/end frames), and omni_reference (up to 12 mixed references with images, videos, audio). Use @image1, @video1, @audio1 syntax in prompts to reference images, videos (mp4/mov), and audio (mp3/wav up to 15 seconds) for creative control. Assign each reference a role so the model knows exactly what comes from where.
Modify characters, objects, or backgrounds inside an existing clip without regenerating the whole scene. Maintains character appearance, product details, typography, and scene consistency across frames. From facial features to clothing patterns, from packaging textures to font clarity—elements remain stable and accurate throughout the video without drift or degradation.
Use short video references (up to 3 clips, 15s total) for motion, camera work, and effects transfer. Extend scenes or apply style transfer with narrative coherence. Stretch a 5-second clip into a longer narrative while keeping characters, lighting, and motion continuous.
Cloth folds, fluids splash, and bodies move with believable weight. The jump from 1.0 to 2.0 brings improvements across several areas: Motion quality: More natural movement, especially for humanoid figures and articulated motion—ideal for product demos, sports, and human-centric storytelling.
Accurately replicates sophisticated camera movements including dolly zooms, tracking shots, and mechanical arm multi-angle follows. Feed in a reference clip and the Seedance 2.0 API will mirror its choreography in a brand-new scene.
The model outputs dual-channel stereo audio with synchronized multi-track output — background music, ambient sound effects, and character dialogue generated in parallel. This includes ambient sounds, sound effects (SFX), and even basic lip-syncing for characters, providing a seamless audiovisual experience.
The integration follows the standard apipass asynchronous jobs flow used by apipass-backed video models.
Sign up and load credits into your wallet. No model subscription, no waitlist.
From the ApiPass dashboard, generate a key and store it as an environment variable.
Choose between text-to-video, first/last-frame image-to-video, or omni-reference multimodal generation depending on how much control you need.
Upload your reference assets and write a structured prompt using the @image1, @video1, @audio1 tag syntax. Set duration (4–15s), aspect ratio, and resolution.
Send your payload to ApiPass's unified /videos/generations route with the seedance-2.0 model identifier.
Retrieve the finished MP4 once rendering completes, then ship it directly into your product, pipeline, or CMS.
These suggestions line up with the adapter guardrails and upstream request schema.
Use either text-only, first/last-frame image-to-video, or multimodal references. Mixing first/last frames with reference media will be rejected by the adapter.
The framing target changes significantly between 16:9, 9:16, 1:1, and adaptive. Decide the canvas first, then describe camera motion and subject placement.
Standard Seedance 2.0 supports 1080p for higher-quality output. Use 480p or 720p for faster iteration and 1080p when final quality matters.
Seedance 2.0 represents a paradigm shift in AI video generation, engineered specifically for coherent multi-shot storytelling. Unlike standard models, Seedance 2.0 automates the creation of logical scene sequences from a single prompt, ensuring visual consistency across shots.
Generate branded video content with precise visual consistency and audio sync for social media, ad campaigns, and product demonstrations.
Lock a character's face, outfit, and voice across multiple scenes using image and audio references, perfect for episodic content and animated series.
Feed an audio track and let the Seedance 2.0 API generate beat-matched visuals with synced character dialogue and ambient SFX.
You upload a still image and Seedance 2.0 animates it. Turn static product shots, illustrations, or photos into living scenes.
Lengthen existing footage, swap backgrounds, or replace objects inside a finished clip without re-shooting or regenerating the entire video.
Skip vendor lock-in and platform queues. ApiPass exposes the full Seedance 2.0 API — text-to-video, first/last frame, and omni-reference modes — through a single, OpenAI-style endpoint you can integrate in minutes.
Only pay for the seconds you actually generate. No monthly minimums, no seat fees, and pricing that scales with your workload from prototype to production.
ApiPass handles the infrastructure — load balancing, retries, and high-concurrency throughput — so your app stays smooth whether you're rendering one hero video or batch-generating thousands of ad variants overnight.
All APIs require authentication via Bearer Token.
Authorization: Bearer
Use the SDK to create tasks, store the returned taskId, and retrieve results for Seedance 2.0.
Install @apipass-dev/apipass-sdk before running these SDK examples.
1npm install @apipass-dev/apipass-sdk
2# or
3pnpm add @apipass-dev/apipass-sdk
4# or
5yarn add @apipass-dev/apipass-sdkKeep your API key in server-side environment variables. The SDK sends it as a Bearer token and adds JSON headers automatically.
client.jobs.createTaskCreate a new Bytedance Seedance 2.0 video generation task through the apipass unified jobs API
Use this example to get started quickly with the selected SDK.
1import { Apipass } from "@apipass-dev/apipass-sdk";
2
3const client = new Apipass({
4 apiKey: process.env.APIPASS_API_KEY!,
5});
6
7const task = await client.jobs.createTask({
8 "model": "bytedance/seedance-2",
9 "callBackUrl": "https://your-domain.com/api/callback",
10 "input": {
11 "prompt": "A cinematic macro shot of a crystal hummingbird hovering above a night-blooming flower. Moonlight refracts through its transparent wings, scattering blue and silver highlights across dew-covered petals.",
12 "duration": 5,
13 "resolution": "720p",
14 "aspect_ratio": "16:9",
15 "generate_audio": false
16 }
17});
18
19console.log(task.data.taskId);apipass keeps the public request envelope stable across asynchronous apipass-backed models. Seedance-specific parameters are passed inside the input object.
1{
2 "model": "bytedance/seedance-2",
3 "callBackUrl": "string (optional)",
4 "input": {
5 "prompt": "string",
6 "first_frame_url": "https://example.com/first.png",
7 "last_frame_url": "https://example.com/last.png",
8 "reference_image_urls": [
9 "https://example.com/ref.png"
10 ],
11 "reference_video_urls": [
12 "https://example.com/ref.mp4"
13 ],
14 "reference_audio_urls": [
15 "https://example.com/ref.mp3"
16 ],
17 "generate_audio": false,
18 "resolution": "480p | 720p | 1080p | 4K",
19 "aspect_ratio": "1:1 | 4:3 | 3:4 | 16:9 | 9:16 | 21:9 | adaptive",
20 "duration": 5
21 },
22 "channel": "auto"
23}modelRequiredstringThe public apipass model name for Bytedance Seedance 2.0.
"bytedance/seedance-2"
callBackUrlOptionalstringOptional callback URL for task completion notifications.
"https://your-domain.com/api/callback"
channelOptionalstringYou may specify the corresponding provider within APIPASS via the channel parameter; these providers handle the actual image and video generation tasks. APIPASS currently offers three provider options:
The default value for the channel parameter is auto. When enabled, APIPASS automatically allocates tasks across available providers based on real-time pricing and stability metrics to balance minimal cost and reliable performance. Retain the default auto value unless you have custom routing requirements.
Available options:
auto
The input object contains the Seedance 2.0 parameters accepted by the apipass. Image-to-video first/last frame inputs and multimodal reference inputs are mutually exclusive.
input.promptRequiredstringText prompt used to generate the video. Must be 3-20000 characters.
"A cinematic macro shot of a crystal hummingbird hovering above a night-blooming flower."
input.first_frame_urlOptionalstringOptional first-frame image URL or asset URL. The adapter also accepts firstFrameUrl, image_url, and image.
"https://example.com/first.png"
input.last_frame_urlOptionalstringOptional last-frame image URL or asset URL. The adapter also accepts lastFrameUrl.
"https://example.com/last.png"
input.reference_image_urlsOptionalarray<string>Enter a list of image URLs.
Single image requirements:
Maximum number of files: The sum of the number of frames at the beginning and end must not exceed 9.
<= 9 items
["https://example.com/ref.png"]
input.reference_video_urlsOptionalarray<string>Enter a list of video URLs.
Single video requirements:
<= 3 items
["https://example.com/ref.mp4"]
input.reference_audio_urlsOptionalarray<string>Enter a list of audio URLs.
Single audio requirements:
<= 3 items
["https://example.com/ref.mp3"]
input.generate_audioOptionalbooleanWhether to generate audio for the video. true: Generate with audio (higher cost) false: Generate without audio
false
input.resolutionOptionalstringRequested video resolution. Standard Seedance 2.0 supports 480p, 720p, and 1080p. The adapter also accepts quality.
Available options:
"720p"
input.aspect_ratioOptionalstringRequested video aspect ratio. The adapter also accepts aspectRatio and defaults to 16:9.
Available options:
"16:9"
input.durationOptionalintegerVideo duration in seconds. The adapter validates values from 4 to 15 and defaults to 5.
5
1{
2 "code": 200,
3 "message": "success",
4 "data": {
5 "taskId": "task_bytedance_1765186743319"
6 }
7}codeStatus code. 200 indicates the apipass task was created successfully.
messageResponse message returned by apipass.
data.taskIdTask identifier used for later status checks.
If callBackUrl is supplied, apipass sends the user callback after the apipass webhook updates the task.
1{
2 "taskId": "task_bytedance_1765186743319",
3 "model": "bytedance/seedance-2",
4 "state": "success",
5 "resultJson": "{\"resultUrls\":[\"https://example.com/generated-seedance-video.mp4\"]}",
6 "failCode": "",
7 "failMsg": ""
8}1{
2 "taskId": "task_bytedance_1765186743319",
3 "model": "bytedance/seedance-2",
4 "state": "fail",
5 "resultJson": "",
6 "failCode": "422",
7 "failMsg": "Validation Error"
8}client.jobs.recordInfoQuery Seedance 2.0 task status and results by task ID through apipass
Use this example to get started quickly with the selected SDK.
1import { Apipass } from "@apipass-dev/apipass-sdk";
2
3const client = new Apipass({
4 apiKey: process.env.APIPASS_API_KEY!,
5});
6
7const result = await client.jobs.recordInfo("task_bytedance_1765186743319");
8
9console.log(result.data.state);
10console.log(result.data.resultJson);1{
2 "code": 200,
3 "message": "success",
4 "data": {
5 "taskId": "task_bytedance_1765186743319",
6 "model": "bytedance/seedance-2",
7 "state": "success",
8 "param": "{\"model\": \"bytedance/seedance-2\", \"input\": {\"prompt\": \"A cinematic crystal hummingbird video.\", \"resolution\": \"720p\", \"aspect_ratio\": \"16:9\", \"duration\": 5}}",
9 "resultJson": {
10 "resultUrls": [
11 "https://cdn.apipass.dev/apipass/results/task_bytedance_1765186743319_0.mp4"
12 ]
13 },
14 "failCode": null,
15 "failMsg": null,
16 "costTime": 0,
17 "completeTime": 1765186800000,
18 "createTime": 1765186743000,
19 "modelRequested": "bytedance/seedance-2",
20 "modelUsed": "bytedance/seedance-2",
21 "fallbackTriggered": false
22 }
23}data.taskIdThe task ID returned by the create endpoint.
data.statePlatform task state. The adapter maps provider state into success, fail, queuing, or generating.
data.resultJson.resultUrlsGenerated video URLs after apipass parses the upstream result and stores normalized output URLs.
data.failCodeFailure code when the task does not complete successfully.
data.failMsgFailure message when the task does not complete successfully.
All APIs require authentication via Bearer Token.
Authorization: Bearer
Create a new Bytedance Seedance 2.0 video generation task through the apipass unified jobs API
apipass keeps the public request envelope stable across asynchronous apipass-backed models. Seedance-specific parameters are passed inside the input object.
1{
2 "model": "bytedance/seedance-2",
3 "callBackUrl": "string (optional)",
4 "input": {
5 "prompt": "string",
6 "first_frame_url": "https://example.com/first.png",
7 "last_frame_url": "https://example.com/last.png",
8 "reference_image_urls": [
9 "https://example.com/ref.png"
10 ],
11 "reference_video_urls": [
12 "https://example.com/ref.mp4"
13 ],
14 "reference_audio_urls": [
15 "https://example.com/ref.mp3"
16 ],
17 "generate_audio": false,
18 "resolution": "480p | 720p | 1080p | 4K",
19 "aspect_ratio": "1:1 | 4:3 | 3:4 | 16:9 | 9:16 | 21:9 | adaptive",
20 "duration": 5
21 },
22 "channel": "auto"
23}modelRequiredstringThe public apipass model name for Bytedance Seedance 2.0.
"bytedance/seedance-2"
callBackUrlOptionalstringOptional callback URL for task completion notifications.
"https://your-domain.com/api/callback"
channelOptionalstringYou may specify the corresponding provider within APIPASS via the channel parameter; these providers handle the actual image and video generation tasks. APIPASS currently offers three provider options:
The default value for the channel parameter is auto. When enabled, APIPASS automatically allocates tasks across available providers based on real-time pricing and stability metrics to balance minimal cost and reliable performance. Retain the default auto value unless you have custom routing requirements.
Available options:
auto
The input object contains the Seedance 2.0 parameters accepted by the apipass. Image-to-video first/last frame inputs and multimodal reference inputs are mutually exclusive.
input.promptRequiredstringText prompt used to generate the video. Must be 3-20000 characters.
"A cinematic macro shot of a crystal hummingbird hovering above a night-blooming flower."
input.first_frame_urlOptionalstringOptional first-frame image URL or asset URL. The adapter also accepts firstFrameUrl, image_url, and image.
"https://example.com/first.png"
input.last_frame_urlOptionalstringOptional last-frame image URL or asset URL. The adapter also accepts lastFrameUrl.
"https://example.com/last.png"
input.reference_image_urlsOptionalarray<string>Enter a list of image URLs.
Single image requirements:
Maximum number of files: The sum of the number of frames at the beginning and end must not exceed 9.
<= 9 items
["https://example.com/ref.png"]
input.reference_video_urlsOptionalarray<string>Enter a list of video URLs.
Single video requirements:
<= 3 items
["https://example.com/ref.mp4"]
input.reference_audio_urlsOptionalarray<string>Enter a list of audio URLs.
Single audio requirements:
<= 3 items
["https://example.com/ref.mp3"]
input.generate_audioOptionalbooleanWhether to generate audio for the video. true: Generate with audio (higher cost) false: Generate without audio
false
input.resolutionOptionalstringRequested video resolution. Standard Seedance 2.0 supports 480p, 720p, and 1080p. The adapter also accepts quality.
Available options:
"720p"
input.aspect_ratioOptionalstringRequested video aspect ratio. The adapter also accepts aspectRatio and defaults to 16:9.
Available options:
"16:9"
input.durationOptionalintegerVideo duration in seconds. The adapter validates values from 4 to 15 and defaults to 5.
5
1curl -X POST "https://api.apipass.dev/api/v1/jobs/createTask" \
2 -H "Content-Type: application/json" \
3 -H "Authorization: Bearer YOUR_API_KEY" \
4 -d '{
5 "model": "bytedance/seedance-2",
6 "callBackUrl": "https://your-domain.com/api/callback",
7 "input": {
8 "prompt": "A cinematic macro shot of a crystal hummingbird hovering above a night-blooming flower. Moonlight refracts through its transparent wings, scattering blue and silver highlights across dew-covered petals.",
9 "duration": 5,
10 "resolution": "720p",
11 "aspect_ratio": "16:9",
12 "generate_audio": false
13 }
14 }'1{
2 "code": 200,
3 "message": "success",
4 "data": {
5 "taskId": "task_bytedance_1765186743319"
6 }
7}codeStatus code. 200 indicates the apipass task was created successfully.
messageResponse message returned by apipass.
data.taskIdTask identifier used for later status checks.
ByteDance
ByteDance
Starting from
0 credits
MiniMax
MiniMax
Starting from
0 credits
ByteDance
ByteDance
Generate cinematic 4K AI videos up to 30 seconds with Seedance 2.5's multimodal power.
Starting from
0 credits
Google Veo 3.1 is the latest state-of-the-art generative video model designed to transform text and image prompts into high-fidelity cinematic visuals. Building upon its predecessors, Veo 3.1 features significant upgrades in prompt adherence, visual realism, and native audio generation, creating 8-second clips with synchronized sound effects and dialogue.
Starting from
0 credits
wan
wan
Starting from
0 credits
Runway
Runway
Starting from
0 credits
Kling
Kling
Transform text prompts and images into high-quality videos with advanced AI. Generate cinematic content with customizable duration, aspect ratio, and audio capabilities.
Starting from
0 credits
MiniMax
MiniMax
Generate high-quality AI videos from text prompts or images with MiniMax's Hailuo 2.3 model. Support for both 6s and 10s durations, 768p and 1080p resolutions, with intelligent prompt optimization.
Starting from
0 credits
grok
grok
Grok Imagine is xAI’s AI video generation model for creating high-quality videos with realistic motion, native audio, and synchronized speech. Grok Imagine 1.5 introduces faster generation speeds, improved motion physics, and stronger visual consistency for modern AI video applications.
Starting from
0 credits
Luma
Luma
Starting from
0 credits
Kling
Kling
Kling v3 video generation model, supports text-to-video and image-to-video, up to 15 seconds, 1080p pro mode
Starting from
0 credits
wan
wan
Generate videos with cinematic consistency by reimagining existing footage through high-fidelity visual transformation and structural control.
Starting from
0 credits
Starting from
0 credits
grok
grok
Starting from
0 credits
kling
kling
Starting from
0 credits
ByteDance
ByteDance
Starting from
0 credits