ByteDance models: APIs and prices

Image generation4

• Seedream 4.0doubao-seedream-4-0-250828

  ByteDance Seedream 4.0 image generation (classic tier).

  per image 10.8Credits
• Seedream 4.5doubao-seedream-4-5-251128

  ByteDance Seedream 4.5 image generation.

  per image 13.5Credits
• Seedream 5.0doubao-seedream-5-0-260128

  ByteDance Seedream 5.0, the image generation flagship.

  per image 11.88Credits
• Seedream 5.0 Prodoubao-seedream-5-0-pro-260628

  ByteDance Seedream 5.0 Pro, the enhanced image generation flagship; reference images are billed per image.

  per image 16.2Credits–32.4Credits·Input image 1.08Credits/image

Video generation9

• Seedance 2.0doubao-seedance-2.0

  ByteDance Seedance 2.0, the video generation flagship (asynchronous task). Text-to-video, image-to-video and multimodal reference; 480p / 720p / 1080p / 4K, 4-15 seconds, 24 fps.

  per second 27.721Creditsfrom
• Seedance 2.0 Fastdoubao-seedance-2.0-fast

  ByteDance Seedance 2.0 Fast, the low-latency video generation tier (asynchronous task).

  per second 22.298Creditsfrom
• Seedance 2.0 Fast Litedoubao-seedance-2.0-fast-lite

  ByteDance Seedance 2.0 fast lite tier: low latency and high value (asynchronous task). Text-to-video, image-to-video and multimodal reference; 720p, 4-15 seconds, 24 fps.

  per second 47.952Credits
• Seedance 2.0 Litedoubao-seedance-2.0-lite

  ByteDance Seedance 2.0 lite tier, high-value output (asynchronous task). Text-to-video, image-to-video and multimodal reference; 720p / 1080p, 4-15 seconds, 24 fps.

  per second 59.616Creditsfrom
• Seedance 2.0 Minidoubao-seedance-2.0-mini

  ByteDance Seedance 2.0 Mini, a light, high-value video generation tier (asynchronous task). Text-to-video, image-to-video and multimodal reference; 480p / 720p, 4-15 seconds, 24 fps.

  per second 13.861Creditsfrom
• Seedance 2.0 Mini Litedoubao-seedance-2.0-mini-lite

  ByteDance Seedance 2.0 Mini lite tier, the lowest cost for light scenarios (asynchronous task). Text-to-video, image-to-video and multimodal reference; 720p, 4-15 seconds, 24 fps.

  per second 29.808Credits
• Seedance 2.5doubao-seedance-2.5

  ByteDance Seedance 2.5, the new video generation flagship (asynchronous task). Text-to-video, image-to-video with a first frame or first and last frames, and multimodal or combined reference with images (up to 30), video and audio; can generate video with sound. 480p / 720p / 1080p, 4-30 seconds, 24 fps, aspect ratios 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16.

  per second 40.352Creditsfrom
• Seedance 2.5 Litedoubao-seedance-2.5-lite

  ByteDance Seedance 2.5 lite tier, high-value output (asynchronous task). Text-to-video, image-to-video with a first frame or first and last frames, and multimodal or combined reference with images (up to 30), video and audio; can generate video with sound. 720p / 1080p, 4-30 seconds, 24 fps, aspect ratios 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16.

  per second 90.72Creditsfrom
• Seedance 2.0 YDseedance2.0-yd

  ByteDance Seedance 2.0 on a separate channel (asynchronous task). Text-to-video and multimodal reference (up to 9 reference images and up to 3 reference audio clips, with roles reference_image / reference_audio); no first frame, first-and-last frame or reference video. A text prompt is required, along with resolution (480p / 720p / 1080p), a specific ratio (16:9 / 4:3 / 1:1 / 3:4 / 9:16 / 21:9; adaptive is not supported) and an integer duration (4-15 seconds), 24 fps.

  per second 27.721Creditsfrom

FAQ

How do I call ByteDance models?

Create an API key on XiaoZhan AIhub, point your SDK's base_url at https://api.dflop.top/v1 and put the model ID from the list above in the model field. The OpenAI, Anthropic and Gemini SDKs all work.

How are ByteDance models billed?

Each model is billed its own way: text models per input and output token, image models per image, video models per second or per token. Prices are in credits (60 credits = ¥1); each model page lists the full price.