If you’re a developer or creator looking to unlock the cutting edge of AI-powered video generation, Seedance 2.0 is a game changer — and this post will show you exactly how to use its curl POST endpoint with Content-Type: application/json headers to get started today.
Behind the scenes, Seedance 2.0 is powered by tech coming out of innovative companies like Apiframe, in the same ecosystem as giants such as ByteDance and popular editor CapCut. Whether you want to turn text prompts, images, or even reference videos into stunning visual content, Seedance 2.0’s unified API endpoint lets you do it all — elegantly and efficiently.
Why Seedance 2.0 Stands Out
In the rapidly evolving landscape of AI video generation, Seedance 2.0 sets itself apart with several key features:
- One single endpoint for text-to-video, image-to-video, and reference-to-video generation. Use of multimodal references with roles — so you can specify separate inputs for style, motion, and sound, allowing highly customizable video outputs. Native synchronized audio synthesized in the same generation pass — no need for separate audio dubbing workflows. Director-style camera movement controlled by natural prompt language, creating dynamic storytelling shots without manual animation. Pricing billed transparently per second of video output, making budgeting straightforward for developers and production teams.
Seedance 2.0 API Overview
The core video generation API is accessible via the POST endpoint:
To track your submitted jobs, you use the GET endpoint with the returned job ID:
Method URL Description GET https://api.apiframe.ai/v2/jobs/id Get status and results of a previously submitted jobSeedance Parameters Explained: seedanceParams
The heart of your POST curl request is the seedanceParams JSON object. This configures your video generation task — specifying modal inputs, video length, style, camera movement, and more.
Some notable keys inside seedanceParams include:
- references: a list of multimodal assets (images, videos, sounds) assigned with distinct roles such as "style," "motion," or "sound" prompt: your natural language description of the video’s narrative and camera movement resolution: output video resolution; sanity check this to match your requirements (e.g., "720p" or "1080p") length: desired duration in seconds generate_audio: whether to synthesize native audio in the generation pass (usually true)
Example curl Request: Seedance 2.0 in Action
Here’s a fully working example curl POST request using Content-Type: application/json that submits a text-to-video generation with a style image reference, motion video reference, and embedded audio synthesis in one go.
curl -X POST https://api.apiframe.ai/v2/videos/generate \ -H "Content-Type: application/json" \ -H "Authorization: Bearer YOUR_API_KEY" \ -d ' "seedanceParams": "prompt": "A serene sunset over a magical forest with gentle camera pan and slow zoom", "length": 12, "resolution": "720p", "generate_audio": true, "references": [ "role": "style", "url": "https://example.com/artistic-style.jpg" , "role": "motion", "url": "https://example.com/motion-pattern.mp4" , "role": "sound", "url": "https://example.com/ambient-sound.wav" ] 'Notes on this example:
- Replace YOUR_API_KEY with your actual API token from Apiframe. The references array uses roles to clearly separate style (how it looks), motion (how camera or elements move), and sound (ambient audio). generate_audio: true tells Seedance to create synchronized audio natively — no post-processing needed. We request a 12-second video at 720p HD resolution, a sensible default unless you want 1080p or ultra HD (which might be billed differently).
Checking Job Status
After submitting your job, you’ll receive a response with a job ID. You can poll the status or retrieve the final output with:
curl -X GET https://api.apiframe.ai/v2/jobs/id \ -H "Authorization: Bearer YOUR_API_KEY"Replace id with the job ID from the response. The API will return JSON containing status codes, progress percentage, and upon completion, URLs for your generated video asset.
Pricing Model: Billed Per Second of Video Output
One important consideration is how pricing is structured.
The Seedance 2.0 platform charges based on actual video output length measured in seconds, not input complexity or reference count. This clear pricing model — billed per second of video output — means you know exactly what your cost will be as you scale production.
For example, a 12-second 720p video will be billed proportionally to those 12 seconds, regardless of whether you used one or multiple reference assets or built elaborate camera how to control camera movement movements in the prompt. This transparency aligns well with toolsets from ByteDance and CapCut, where creators expect straightforward, usage-based billing.
Integrations and Ecosystem: Apiframe, ByteDance, CapCut
It’s worth noting that Seedance 2.0 is part of an ecosystem driven by Apiframe, whose API-first approach powers fast integration for developers and creators.
In the same space, ByteDance and its popular editing platform CapCut push forward the boundaries of video creation and consumption. Seedance 2.0’s multimodal, prompt-driven, native audio approach fits neatly into this ecosystem, enabling creators to generate unique, dynamic content easily embedded into modern social and ad-tech workflows.
Best Practices and Tips
Always sanity-check your defaults: For example, make sure your resolution matches your final destination. 720p is a good balance, but if you want crisp YouTube-ready videos, 1080p might be better. Explicitly set generate_audio: true if you want synchronized soundtracks embedded. Carefully name your multimodal references with the correct role to instruct Seedance which URL contributes what aspect (style, motion, sound). Test with short clips and short video duration to save costs during experimentation. Use the status GET endpoint to programmatically track and integrate video generation into async pipelines.Summary
Seedance 2.0 offers one of the most developer-friendly, powerful APIs for AI video synthesis today — leveraging a single endpoint that flexibly handles text, image, and video inputs with clear roles, synchronized audio generation, and director-level camera motion controlled by natural prompt language.
Using the curl POST example in this guide with Content-Type: application/json, you can start building video creative workflows that integrate seamlessly with the broader Apiframe ecosystem and are competitively priced (billed by output seconds) — just like solutions from ByteDance and CapCut.
Get your API key, try the example, and begin turning your ideas into dynamic video assets now!