Seedance Lite usually refers to Seedance 1.0 Lite, but API access and specifications depend on the provider. Together AI lists ByteDance/Seedance-1.0-lite at $0.14 per video and documents 720p, 5-second clips. fal lists Seedance 1.0 Lite with a separate image-to-video endpoint and configurable settings, including 480p, 720p, or 1080p and 2–12-second durations. These are provider-specific offers—not one universal model ID, endpoint, or set of limits. (Together AI)
What does “Seedance Lite” mean?
The specific model name shown in the API listings is Seedance 1.0 Lite. ByteDance’s Seed site describes Seedance 1.0 as a text- and image-based, multi-shot video model, but the cited first-party model page does not identify a Lite variant. The Lite name and its API availability are therefore best treated as provider-listed offerings, with each provider’s documentation governing what you can call. (ByteDance Seed)
A third-party CometAPI page describes Seedance 1.0 Lite as a streamlined Seedance 1.0 version and claims it balances speed, cost, and quality. It provides no comparative benchmarks, so those claims do not establish how Lite performs through a particular API provider. Don’t treat “Lite” as proof that a generation will be faster, cheaper, or lower quality without comparing the provider’s actual terms and results. (CometAPI)
What can it do, and what limits are documented?
The provider references support these specific conclusions:
- Together AI: Its listing identifies
ByteDance/Seedance-1.0-lite, thePOST https://api.together.ai/v2/videosendpoint, and a model card describing 720p, 5-second clips. Its documented cURL request includes a model ID and prompt. The listed price is $0.14 per video on Together AI; don’t assume that price applies through other providers or to different model settings. (Together AI) - fal: Its
fal-ai/bytedance/seedance/v1/lite/image-to-videoendpoint requires bothpromptandimage_url. The schema lists aspect-ratio choices of 21:9, 16:9, 4:3, 1:1, 3:4, 9:16, orauto; resolutions of 480p, 720p, or 1080p; and durations from 2 to 12 seconds. These are fal’s settings for that image-to-video endpoint, not a universal specification for every Seedance Lite integration. (fal)
The CometAPI capture also says text-to-video and image-to-video, but it does not provide enough technical detail to establish a common set of supported modes or limits across providers. Check the exact provider’s endpoint documentation for required inputs, supported settings, response handling, and operational limits. (CometAPI)
When should you choose Seedance Lite?
Choose based on the documented option on the provider you plan to use, rather than generic claims about the Lite label:
- Consider Together AI’s listing if its documented 720p, 5-second output fits your job and its listed price of $0.14 per video fits your budget. That price is specific to Together AI, not evidence of what another provider charges. (Together AI)
- Consider fal’s image-to-video endpoint if you need image input or the option to select among that endpoint’s documented aspect ratios, resolutions, and durations. Confirm current account pricing separately; the cited endpoint documentation lists configuration, not a price. (fal)
- Investigate Seedance 2.5 instead if your requirement is specifically a longer narrative clip with reference control or editing. ByteDance describes Seedance 2.5 as built for 30-second storytelling with reference control and editing capabilities. That is not a like-for-like Lite comparison, and this evidence does not establish that 2.5 is faster, better, or more cost-effective for your job. (ByteDance Seedance 2.5)
If quality, latency, or total generation cost will decide the choice, compare the provider options you can access using comparable prompts and settings. The sources cited here do not provide a like-for-like Lite-versus-other-model benchmark.
Production API: what to check before integrating
API access is documented by specific providers, but their identifiers and request formats differ. Together AI lists ByteDance/Seedance-1.0-lite with POST https://api.together.ai/v2/videos and provides a cURL request example with bearer-token authentication. Its documented example body contains model and prompt fields. fal documents fal-ai/bytedance/seedance/v1/lite/image-to-video, whose example includes both a prompt and an image URL. Use the model ID, inputs, and authentication documented for the provider you choose; do not copy identifiers or request fields from another service. (Together AI; fal)
Together AI documents video generation as asynchronous: creating a video returns a job ID; the client then retrieves job status, and the completed response provides the generated video at outputs.video_url. Its video-generation guide describes this workflow. Confirm the current model-specific behavior and any account limits before relying on it in production.
Before integrating either provider, confirm its current input schema, completion and output-retrieval flow, error handling, account access, operational limits, and applicable price. The documented routes establish provider-specific API access; they do not establish universal service levels, rate limits, regional availability, or production guarantees.
Bottom line: Seedance 1.0 Lite is available through the cited Together AI and fal API listings, but there is no single provider-independent Lite API specification. Match the provider’s documented model ID, input mode, output settings, response flow, and billing terms to your workload, and avoid treating unmeasured speed or quality claims as selection facts.
Sources
ByteDance Seedance 1.0 Lite — Together AI
Bytedance Seedance V1 Lite API — fal