← All models

Seedance 1.0

Published Publisher ByteDance

Model Overview

Seedance 1.0 is a foundation model for video generation from ByteDance's Seed team. It was officially released on 2025-06-11, and its technical report was made public on the same day. (seed.bytedance.com)

Capabilities and Uses

  • Supports text-to-video and image-to-video generation, with video output up to 1080p.
  • Supports native multi-shot storytelling; the official demonstration showed the ability to switch between 2–3 shots in a 10-second video.
  • Can be used to create film and television content, advertisements, animation, and other stylized videos. The official announcement emphasized improvements in motion stability, instruction following, and visual quality. (seed.bytedance.com)

Limitations

  • Generated results may still have motion, structural, or detail issues and are not guaranteed to fully match the prompt; specific features and output specifications depend on the platform and version used.
  • The official report lists audio input and output as areas for future exploration; Seedance 1.0 Pro, as listed in the Volcengine Ark documentation, does not support generating videos with sound. (seed.bytedance.com)
  • The 41.4-second time to generate a 5-second, 1080p video in the official report was a test result based on L20, not a speed guarantee for all environments. (seed.bytedance.com)

Pricing

The following are official prices for specific Seedance 1.0 Pro service versions and should not be regarded as uniform pricing for all Seedance 1.0 products or versions; all were checked on 2026-11-18.

  • Mainland China: Volcengine Ark, Chinese yuan (CNY). Seedance 1.0 Pro online inference costs 15.00 yuan/million tokens, and offline inference costs 7.50 yuan/million tokens. Video generation is billed based on the token usage of the generated video; video generation models do not count input tokens, so input tokens are 0. Actual costs depend on output specifications and usage. Sources: Volcengine Ark pricing; video generation billing information. (docs.volcengine.com)
  • International: BytePlus ModelArk, US dollars (USD). seedance-1-0-pro-250528 online inference costs 2.50 US dollars/million tokens, and offline inference costs 1.25 US dollars/million tokens; text-to-video and image-to-video use the same unit price. Billing is based on video generation token usage, with actual usage determined by usage.completion_tokens returned by the API; offline inference is suitable for tasks that can tolerate higher latency. Sources: ModelArk pricing; model documentation. (docs.byteplus.com)

The domestic and international prices are service prices for different platforms and should not be treated as the same plan with directly convertible prices. The Seedance 1.0 listed in the input does not specify a particular API version, so the Pro prices above are provided only as a reference for the stated version.

Official Sources