- 🏢 ByteDance Seed's flagship video generation family, released March 2026.
- 🆕 Unified multimodal architecture accepting text, image, audio and video inputs.
- 💬 Native joint audio-video generation, so sound is produced with the picture.
- 🎯 Director-level control over performance, lighting, shadow and camera movement.
- 🔧 Emphasis on motion stability and scene consistency across a shot.
- ⚡ Family also includes Fast and Mini variants, plus image- and reference-conditioned endpoints.
- 🌐 This entry is the text-to-video endpoint of the 2.0 family.
ByteDance is a Chinese internet technology company headquartered in Beijing's Haidian district, with an associated variable-interest entity incorporated in the Cayman Islands. Best known as a consumer internet operator at enormous scale, it has also become a significant…
Explore 18 more models by ByteDance →Seedance 2.0 is the text-to-video member of ByteDance Seed's flagship video family. According to Seed's own model page, it adopts a unified multimodal audio-video joint generation architecture that accepts text, image, audio and video inputs, and emphasizes motion stability plus creator control over performance, lighting, shadow and camera movement. In practice, a prompt can specify camera moves and staging, and the model generates video with synchronized audio rather than a silent clip.
Against its predecessor Seedance 1.5 Pro, the documented change is architectural rather than incremental: Seed positions 2.0 as a joint audio-video model with multimodal conditioning across text, image, audio and video, where earlier Seedance releases were framed as video generators driven by text or a starting image. Seed's stated priorities for this generation are motion stability and director-style control of performance and lighting.
The 2.0 release ships as a tier system in this catalog. Seedance 2.0 Fast and the later Seedance 2.0 Mini are smaller-tier variants, while Seedance 2.0 image-to-video and Seedance 2.0 R2V expose the same generation through image and multi-reference conditioning.
For newer work, Seedance 2.5 succeeds this checkpoint in the same family, alongside its Seedance 2.5 image-to-video counterpart. Seedream V5 Pro is the image model in the same provider family.
This About section is AI-generated from public sources via VeniceStats + Venice inference, with no human editing. It may contain inaccuracies.
Daily usage · Seedance 2.0 Basic · Organic sellers
| Seller | Health | Rep↓ | Sales | Video | Routing | $ / s | $ / video |
|---|---|---|---|---|---|---|---|
| ▲ Apex Ant 0x73b4…e736 | — | 81.71 | $10.98 | 2.1 min | #2 | $0.0851 | — |
| Venice.ai Proxy 0x1f22…18c9 | failingupstream error · 3.7 h · stale | 86.84 | $95.39 | 17.2 min | #1 | $0.104 | — |
"Best price" and the seller table are live Antseed catalog data (advertised $/1M tokens — or $ per generated image for unit-billed image models — not settled amounts). Reputation = buyer trust score (0-100, the Antseed SDK's own formula). "Routing" = the SDK's default buyer routing (what the VPR desktop app ships with): a trust ≥ 60 gate on the effective reputation, then cheapest-first among routable sellers; live failover state (per-peer cooldowns) is buyer-side runtime and not included. Model knowledge (TLDR, provider, About) via the VeniceStats enrichment layer. Advertised catalog, not the model used in any specific purchase. "Usage on Antseed" counts only settlements whose buyers share the per-model split on-chain (metadata v2/v3, opt-in), so every usage figure is a lower bound.