Dreamina API v1

February 23, 2026 (March 4, 2026)

This is the experimental API for Dreamina by ByteDance.

Dreamina creates AI-generated videos and images using models from ByteDance’s Seed family.

Video generation models (US region):

  • seedance-2.0 — Latest model, supports text-to-video, image-to-video (first frame), and end frame modes. 4-12 second duration.
  • seedance-1.5-pro — Pro model with 5/10/12 second durations.
  • seedance-1.0-mini — Supports text-to-video, image-to-video, and multi-frame keyframe mode. 5/10 second duration.

Image generation models (US region):

  • seedream-5.0-lite — Latest Seedream model, 2K/4K resolution, up to 6 reference images, auto ratio.
  • seedream-4.6 — Latest stable model, 2K/4K resolution, up to 6 reference images, auto ratio.
  • seedream-4.5 — High quality, 2K/4K resolution, up to 6 reference images, auto ratio.
  • seedream-4.1 — 2K/4K resolution, up to 6 reference images, auto ratio.
  • seedream-4.0 — Balanced quality/speed, 2K/4K resolution, up to 6 reference images, auto ratio.
  • nano-banana — Fast generation, 1K resolution, up to 3 reference images.
  • seedream-3.0 — 1K/2K resolution, max 1 reference image, no auto ratio.

Image upscale to 2K/4K/8K resolution via POST /images/upscale.

Cost estimate (Dreamina Advanced subscription):

  • Monthly plan ($40/mo, 30K credits): a single 12-second Seedance 2.0 video costs 72 credits — approximately $0.095 per generation.
  • Yearly plan ($335/yr, 30K credits/mo): cost drops to under $0.07 per generation.
  • Every account also receives 2 free generations per day, further reducing effective cost.
  • Both plans include free and unlimited Seedream 4.0 image generation, including 2K resolution.

⚙️ Setup Dreamina

📦 Postman collection (March 4, 2026)

🤖 LLM-friendly API spec Feed this to your LLM to build integrations

Examples:

Developer Community:

Table of contents