Wan 3.0 Prime
alibaba/wan-3.0-prime/text-to-videowan-3-0-prime-text-to-video- 🏢 Alibaba's premium text-to-video tier in the third-generation Wan family.
- 🆕 Released August 2026 alongside image-to-video and reference-to-video Prime variants.
- 📏 Alibaba says Wan 3.0 generates up to 30 seconds in one pass.
- 🌐 Inputs can include text, images, audio, video and documents.
- 🔧 One all-in-one model spans text, image and reference-driven video tasks.
- 🎯 Catalog positions it for photorealism and subject coherence across frames.
- ⚡ Served through Alibaba Cloud Model Studio's video generation APIs.
- 📚 Follows the earlier Wan 2.7 and Wan 2.6 video generations.
Alibaba Group is a Chinese multinational technology company founded in 1999 and headquartered in Hangzhou, Zhejiang. Originally built around e-commerce and cloud computing, Alibaba has become one of the most prolific contributors to open-weight AI research, developing the Qwen…
Explore 51 more models by Alibaba Group →Wan 3.0 Prime is the premium text-to-video tier of Alibaba's third-generation Wan video family, listed here alongside Wan 3.0 Prime for image-to-video and Wan 3.0 Prime Reference. In this catalog it targets photorealistic clips where a subject, prop or setting must stay stable as a shot develops.
The structural change comes from the base generation, Wan 3.0. Alibaba describes Wan 3.0 as an all-in-one system that handles text-to-video, image-to-video and reference-driven generation inside a single model, and that extends single-pass output to about 30 seconds. Input coverage widens too: beyond text and stills, Alibaba says generation can be conditioned on audio, existing video and documents, so a brief or a deck can seed a clip.
Compared with Wan 2.7 and Wan 2.6, the differences are scope and interface. Wan 2.7's API reference had already reworked control surfaces, replacing a dedicated shot-type parameter with prompt-level shot description and swapping fixed pixel sizes for a resolution tier plus aspect ratio. Earlier 2.1 through 2.6 image-to-video work was documented as separate first-frame endpoints. Wan 3.0 consolidates those task-specific paths and lengthens the usable single take.
Prime sits above the standard Wan 3.0 entry as the higher-fidelity option for the same text-to-video task, accessed through Alibaba Cloud Model Studio's video generation service.
This About section is AI-generated from public sources via VeniceStats + Venice inference, with no human editing. It may contain inaccuracies.
Daily usage · Wan 3.0 Prime · Organic sellers
| Seller | Health | Rep↓ | Sales | Video | Routing | $ / s | $ / video |
|---|---|---|---|---|---|---|---|
| Open Forge 0x1d90…b0aa | healthy | 91.11 | $284.92 | 34 min | #1 | $0.14 | — |
| ▲ Apex Ant 0x73b4…e736 | — | 81.71 | $1.64 | 20 s | #2 | $0.0819 | — |
| OpenCheap 0x0ee1…6c41 | — | 12.43 | — | — | gated | $0.1378 | — |
"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.