Seedance 2.0 Mini

bytedance/seedance-2-0-mini
OfficialVideo-to-VideoImage-to-VideoText-to-VideoAudio-to-Video

Seedance 2.0-mini offers rapid video generation. It generates 4-15 second videos with text prompts, supports various aspect ratios, audio generation, and enhanced web search capabilities.Reference image width must be between 300 px and 6,000 px.

Read Me

ByteDance Seedance 2.0 Mini API

Introduction

Seedance 2.0 Mini is a cost-efficient multimodal video model in the Seedance 2.0 family. It supports text, image, video, and audio references, allowing creators to guide characters, scenes, motion, camera work, pacing, and sound within one generation task.

Compared with the standard version, Seedance 2.0 Mini is better suited for frequent testing, batch production, and short-form content. It supports 480p and 720p output, making it a practical choice for social media, ad testing, e-commerce, and creative previews.

Key Features

  • Cost-Efficient Multimodal Generation: Seedance 2.0 Mini keeps the core multimodal controls of the Seedance 2.0 series while reducing generation costs. It is useful for testing several prompts, formats, hooks, or creative directions before choosing a final version.

  • Image, Video, and Audio References: Use images to guide characters, products, or scenes. Use videos to guide motion, camera movement, pacing, or effects. Audio can provide dialogue, music, rhythm, or environmental sound.

  • Native Audio-Video Generation: The model can generate visuals and sound together. Dialogue, ambient sound, action effects, and music can follow the timing and movement of the video.

  • Reference-Guided Motion and Camera Work: A reference video can guide movement, camera behavior, editing rhythm, or visual effects. You can keep the original motion structure while changing the subject, product, or setting.

  • Short-Form Video Production: Seedance 2.0 Mini supports adjustable duration and multiple aspect ratios for short-form production. Its 480p and 720p options are designed for previews, social content, and frequent publishing.

  • Cost-Efficient Reference Video Generation: Using a reference video lowers the generation price per second. This is useful for reusing existing motion, camera pacing, or visual structure in new creative versions.

How to Use

  1. Add reference files — Upload images, videos, or audio to guide the subject, motion, camera, or sound.

  2. Write your prompt — Describe the video and explain what each reference should control.

  3. Choose output settings — Select resolution, aspect ratio, duration, audio, and watermark options.

  4. Submit the request — Generate the video and review motion, consistency, and audio sync.

Parameters

Please refer to API doc for details.

Pricing

Pricing depends on resolution, duration, and reference-video usage. Choose 480p for low-cost testing and 720p for publish-ready social and marketing content.

Please refer to pricing list for details.

Use Cases

  • E-Commerce Video Variants: Create several product videos from the same source image, using different backgrounds, selling points, actions, and pacing.

  • Paid Ad Testing: Test different hooks, product angles, formats, and visual structures before increasing campaign spend.

  • UGC-Style Content: Generate short videos for unboxings, product reviews, tutorials, and lifestyle scenes using character, product, and voice references.

  • Motion and Camera Previews: Use reference videos to test actions, camera movement, transitions, and pacing before producing a higher-resolution final version.

  • Batch Social Content: Produce multiple vertical videos for social posts, campaigns, events, and regularly updated content channels.

  • Creative Remixes: Reuse motion, rhythm, or visual effects while replacing characters, products, backgrounds, or story details.

Prompt Tips

  • Keep Each Video Focused: Build the video around one main action, product benefit, or short story. Simple goals usually produce more stable results than several competing ideas.

  • Give Each Reference a Clear Role: Explain which file controls the subject, motion, camera, or audio. Clear assignments reduce confusion between different reference materials.

  • Describe Motion in Order: Write actions in the order they happen, from the opening movement to the final position. This helps the model keep the sequence clear.

  • Separate Subject and Camera Movement: Describe what the subject does separately from how the camera moves. This makes both types of motion easier to follow.

  • Match Complexity to Resolution: Use 480p for motion tests and layout previews. Choose 720p when facial detail, product shape, or background clarity matters more.

  • Use Audio as a Timing Guide: Explain when dialogue, music changes, or sound effects should happen. Connect each audio event to a specific action or shot.

  • Generate Variations Before Selecting: Use the lower cost to test several hooks, camera moves, or pacing options, then choose the strongest version.

Notes

  • prompt, ratio, and duration are required for every generation request.
  • Image, video, and audio references provide optional creative control inputs.
  • Available output resolutions are limited to 480p and 720p only.
  • Pricing changes according to resolution, duration, and reference-video usage for each request.
  • Default settings use five seconds, 480p, 9:16, and enabled audio.
  • Review motion, character consistency, audio synchronization, and brand details carefully.
  • Seedance 2.0: Create higher-resolution commercial videos with more detailed motion, sound, and multimodal control.

  • Seedance 2.0 Fast: Balance generation speed and visual quality for rapid testing and batch video production.