Seedance2-5
Seedance2-5 — AI video generator with controllable 30s videos and up to 50 multimodal references.
Last verified:
What is Seedance2-5?
Seedance2-5 — Seedance 2.5 is an AI video generation model that creates up to 30 seconds of continuous video from multimodal inputs including text, images, video, audio, and 3D models. It supports up to 50 references and offers precise, frame-by-frame creative control for cinematic, production-ready video generation.
Seedance2-5 pricing
Pricing model: Paid
Limited-time sale: 50% off first annual payment (specific pricing details on pricing page)
Seedance2-5 pros
- Generates up to 30 seconds of continuous video in a single pass without stitching
- Supports up to 50 multimodal references to define characters, products, styles, and actions
- Enables second-by-second creative control for precise edits without regenerating entire videos
- Official model variant supports media featuring real people
Seedance2-5 cons
- Maximum video length limited to 30 seconds per generation
- Requires structured multimodal reference inputs and detailed prompts for optimal results
- Multiple pricing tiers (Lite, Mini, Pro, Fast) suggest variable quality/speed trade-offs
Frequently asked questions about Seedance2-5
What is the maximum video length Seedance 2.5 can generate?
Up to 30 seconds of continuous video in a single pass without stitching.
What types of inputs can I use as references?
Up to 50 multimodal references including text, images, video, audio, and untextured 3D models.
Does it support videos with real people?
Yes, the official Seedance 2.5 model supports media featuring real people.
Can I edit specific parts of a video without regenerating it entirely?
Yes, Seedance 2.5 enables precise, targeted edits with second-by-second control while keeping scene structure consistent.