seedance-2.5-liteSeedance 2.5 Lite is the budget tier of ByteDance Seedance 2.5 on apimodels.app, built for two jobs: text-to-video, and video guided by reference images. Every clip costs a flat $2.50, whether you ask for 15 seconds or 30 seconds, and you are charged only when a video is delivered. For comparison, the same 30-second 720p clip on seedance-2.5 bills at $0.270 per second, about $8.10, and 15 seconds about $4.05. The trade is a narrower set of options. duration is 15 (the default) or 30 seconds; anything else returns a 400 at create time instead of being stretched and billed. Output is 720p only, in one of six aspect ratios: 16:9 (default), 9:16, 1:1, 4:3, 3:4 or 21:9. A prompt of up to 3,000 characters is required. You can add up to 9 reference images in reference_image_urls (images is accepted too) as public HTTPS URLs or data URIs, in JPG or PNG; they guide characters, products and style, but they are references, not a first frame, so the opening shot is not pinned to any one image. Name them in the prompt, for example "the bottle from image 1 on the table from image 2". Audio is always generated in the same pass, and there is no switch to turn it off. This tier has no first or last frame, no reference video, no reference audio and no other resolution; each of those returns a clear 400, and seedance-2.5 is the model for them. Expect several minutes per clip: in our own test on 2026-10-22, a 30-second 720p clip came back in 401 seconds (1280x720, 24 fps, H.264 with an AAC soundtrack, no watermark). Some generations fail on this budget tier; a failed job is refunded in full, so retry it or move the job to seedance-2.5 when it must succeed. Call it with model seedance-2.5-lite on POST /api/v1/video/generations, then poll GET /api/v1/video/generations?task_id=xxx every 10-15 seconds.
References keep a character, product or style; they are not a first frame. Up to 10MB each.
Generated video will appear here
Provide URLs and click Generate
15 or 30 seconds at 720p, the same price either way. The same 30-second 720p clip on seedance-2.5 bills at $0.270/s, about $8.10. Only successful generations are charged
Write a prompt alone, or add up to 9 JPG or PNG images to keep a character, product or style. References guide the content; they are not a first frame
duration is 15 (default) or 30 and resolution is 720p only. Any other length or resolution returns a 400 at create time instead of being rewritten and billed
16:9 (default), 9:16, 1:1, 4:3, 3:4 and 21:9: landscape, vertical, square and ultra-wide from one model id
Every clip comes back with an AAC soundtrack generated in the same pass. There is no silent mode; seedance-2.5 accepts generate_audio: false
Delivery is not guaranteed on this tier, and failed jobs are refunded automatically. For first/last frames, video or audio references, other lengths, or work that must succeed, use seedance-2.5
Seedance 2.5 Lite is a Video Generation API provided by ByteDance. Seedance 2.5 Lite is the budget tier of ByteDance Seedance 2.5 on apimodels.app, built for two jobs: text-to-video, and video guided by reference images. Every clip costs a flat $2.50, whether you ask for 15 seconds or 30 seconds, and you are charged only when a video is delivered. For comparison, the same 30-second 720p clip on seedance-2.5 bills at $0.270 per second, about $8.10, and 15 seconds about $4.05. The trade is a narrower set of options. duration is 15 (the default) or 30 seconds; anything else returns a 400 at create time instead of being stretched and billed. Output is 720p only, in one of six aspect ratios: 16:9 (default), 9:16, 1:1, 4:3, 3:4 or 21:9. A prompt of up to 3,000 characters is required. You can add up to 9 reference images in reference_image_urls (images is accepted too) as public HTTPS URLs or data URIs, in JPG or PNG; they guide characters, products and style, but they are references, not a first frame, so the opening shot is not pinned to any one image. Name them in the prompt, for example "the bottle from image 1 on the table from image 2". Audio is always generated in the same pass, and there is no switch to turn it off. This tier has no first or last frame, no reference video, no reference audio and no other resolution; each of those returns a clear 400, and seedance-2.5 is the model for them. Expect several minutes per clip: in our own test on 2026-10-22, a 30-second 720p clip came back in 401 seconds (1280x720, 24 fps, H.264 with an AAC soundtrack, no watermark). Some generations fail on this budget tier; a failed job is refunded in full, so retry it or move the job to seedance-2.5 when it must succeed. Call it with model seedance-2.5-lite on POST /api/v1/video/generations, then poll GET /api/v1/video/generations?task_id=xxx every 10-15 seconds. Through APIMODELS platform, you can access this model via a unified API with transparent pay-as-you-go pricing. Current pricing: 15s · 720p: $2.5, 30s · 720p: $2.5.
Related model: Seedance 2.5 — for first/last frames, reference video or audio, editing and extension, 480p or any length from 4 to 30 seconds — 480p $0.120/s, 720p $0.270/s
Related model: Seedance 2.0 — when you need 1080p or a first frame, 4-15 seconds — from $0.092/s
Quickly generate brand promotion videos for ad campaigns and social media marketing.
Create compelling short-form video content for platforms like TikTok, Instagram, and YouTube.
Generate product feature demonstrations and tutorials to improve user conversion.
Produce course explanations, knowledge explainers, and training videos at an affordable price.
Seedance 2.5 Lite is available through APIMODELS at: 15s · 720p: $2.5, 30s · 720p: $2.5. Billing is pay-as-you-go — you only pay for what you generate.
Sign up at APIMODELS, get your API key, and call our unified API endpoint. We provide detailed API documentation with code examples in cURL, Python, and Node.js.
APIMODELS offers the same Seedance 2.5 Lite model through our aggregation platform. We provide a unified API interface so you do not need separate accounts for each provider - one API key to access all models.
Seedance 2.5 Lite is the budget tier of ByteDance Seedance 2.5 on apimodels.app, for text-to-video and video guided by up to 9 reference images. It makes 15- or 30-second clips at 720p with audio generated in the same pass, for a flat $2.50 per clip. Call it with the model id seedance-2.5-lite on POST /api/v1/video/generations.
A flat $2.50 per clip, whether you ask for 15 or 30 seconds. Aspect ratio and the number of reference images do not change the price, and only successful generations are charged; failed jobs are refunded. For comparison, the same 720p clip on seedance-2.5 bills at $0.270 per second: about $4.05 for 15 seconds and $8.10 for 30.
Lite trades options for price. It does text-to-video and reference-image video only, at 720p, in 15 or 30 seconds, with audio always on. seedance-2.5 adds first and last frames, reference video and audio, video editing and extension, 480p, any length from 4 to 30 seconds and silent output, billed per second. Pick Lite when a fixed 720p clip at a known price is enough.
Send up to 9 JPG or PNG images in reference_image_urls (images also works) as public HTTPS URLs or data URIs, and name them by number in the prompt, for example: the serum bottle from image 1 on the wooden table from image 2. They guide characters, products and style; they are not a first frame, so the opening shot is not pinned to any one image.
Plan on several minutes per clip; in our own test on 2026-10-22, a 30-second 720p clip came back in 401 seconds. Poll GET /api/v1/video/generations?task_id= every 10-15 seconds or pass callback_url. Some jobs fail on this budget tier; a failed job is refunded in full, so retry it, or use seedance-2.5 for work that must succeed.
It has no first or last frame, no reference video, no reference audio, no silent output and no resolution other than 720p, and it takes at most 9 images, JPG or PNG only. Each of those requests returns a 400 at create time, so nothing is billed. Use seedance-2.5 when you need any of them.
On APIMODELS, Seedance 2.5 Lite runs alongside 60+ models on one API key and one balance, so choosing is about fit, not lock-in. It supports Text to Video, Up to 9 Reference Images, 15s or 30s, 720p, 6 Aspect Ratios, Audio Always On, $2.50 Flat per Clip, and you can weigh it on price and capability against other Video Generation models, then switch by changing a single model-name string — no new account or integration. Browse every Video Generation option with live pricing at apimodels.app/models.
Seedance 2.5 Lite supports: Text to Video, Up to 9 Reference Images, 15s or 30s, 720p, 6 Aspect Ratios, Audio Always On, $2.50 Flat per Clip. See the APIMODELS docs for full parameters and call examples.
Yes. APIMODELS exposes Seedance 2.5 Lite through a single unified API and one key — no separate provider accounts, and no need to handle each provider's regional network access yourself.
We support Stripe (Visa, Mastercard, and other international cards) and Alipay. Credits are available instantly after payment.