Wan (Wan 2.1)
Alibaba's open-weight video AI — Wan 2.2 is the current flagship.
Overview
Alibaba's Wan video-model family is led by the closed, API-only Wan 2.7 (April 2026, native audio sync and first/last-frame control); open weights remain at Wan 2.2 (Apache 2.0 on GitHub), making Wan 2.1 — this entry's original namesake — a superseded but still-downloadable prior generation.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 1 Aug 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- Daily bonus credits
- Text-to-video & image-to-video
- Community licence
- ~60 videos/month
- Priority queue
- Commercial use rights
- ~240 videos/month
- Faster generation
- Commercial use rights
Use cases
What you can produce with Wan (Wan 2.1)
- 5-second text-to-video clips at 480p or 720p resolution
- Image-to-video animations from a single still photograph
- First-and-last-frame interpolation videos for controlled scene transitions
- Video editing and stylisation via VACE model variant
- Bilingual video content with legible Chinese and English text rendered inside frames
- Locally hosted inference pipeline on consumer NVIDIA GPUs (RTX 3080 and above)
- Fine-tuned derivative models via Apache 2.0 open weights
ASEAN Perspective
Wan (Wan 2.1) in Southeast Asia
Wan 2.1 holds notable relevance for the ASEAN region: it is developed by Alibaba, whose Cloud division has its deepest Southeast Asia footprint in Singapore and Malaysia (five data centres in Malaysia alone as of mid-2025), and it natively supports Chinese text rendering — useful for Chinese-language content creators across Singapore, Malaysia, and Vietnam. The model is actively deployed by ASEAN-facing AI platforms such as GladCube's Dra Vis and TabSpace.ai via Alibaba Cloud's Model Studio. Its zero-cost local deployment path is a significant equaliser for developers and SMEs in cost-sensitive markets like Indonesia, the Philippines, and Vietnam, where commercial video AI subscriptions are prohibitively expensive for most creators.
Wan 2.1's February 2025 splash — an 86.22% VBench score that beat OpenAI's Sora at launch — was real, but two things have moved since. Alibaba shipped Wan 2.2 (still Apache 2.0, still free, now the open-weights flagship on GitHub with 16,000-plus stars), and OpenAI killed Sora, shutting the consumer app in April 2026 after burning roughly $1 million a day on a shrinking user base. The field is now Kling 3.0, Seedance 2.0, and Veo 3.1, none of which Wan beats on raw output quality.
What Wan still wins on is the deal: free weights, an Apache 2.0 licence with no royalties, and a 1.3B variant that runs on an 8GB consumer GPU. Alibaba's later 2.5 and 2.6 releases went closed, API-only — a sign the company holds back its best work once it can charge for it. For builders who need open weights, Wan remains the default; for raw quality, it's no longer the benchmark leader.
What people say
OpenAI shut Sora down in April 2026 — the app first, the API by September — after its user base collapsed from roughly a million to under 500,000 and the compute bill hit about $1 million a day. That single fact dates most of the Wan 2.1 commentary written back in 2025, which spent a lot of words on "Wan beat Sora on VBench." True at the time (86.22% vs 84.28%), but the competitor it beat doesn't exist anymore.
What's more relevant in July 2026: Alibaba kept iterating. Wan 2.2 followed 2.1 as another fully open, Apache 2.0 release — both models live on GitHub under Wan-Video with over 16,000 stars each — while 2.5 and 2.6 shipped closed and API-only, available only through Alibaba Cloud or third-party hosts. That's the pattern worth watching: Alibaba open-sources the generation the market has already caught up to and keeps its frontier model closed.
For anyone who actually wants to run video generation locally, the pitch hasn't changed much: free weights, no royalties, a 1.3B model that fits on an 8GB card, and full commercial rights out of the box. Hosted access through the official wan.video app runs on a credit subscription — $5/month for 300 credits, $20/month for 1,200 — which undercuts Runway or Kling's paid tiers by a wide margin. Generation speed on consumer hardware is still slow, and the 14B model wants a real GPU. Against Kling 3.0 and Seedance 2.0 on pure image quality, Wan loses; on cost and openness, nothing else comes close.
Summary of public user & expert reviews, compiled by RECATOOLS.
Notable facts
- Wan 2.1 is the only open-source model to rank in the global top five on the VBench video benchmark leaderboard at its February 2025 launch.
- The 1.3B lightweight variant requires just 8.19 GB of VRAM — enough to run on a standard RTX 3080 or 4070 gaming GPU.
- The technical paper has more than 60 named co-authors, making it one of the largest team-authored AI model reports of 2025.
- Wan 2.1 was the first video generation model capable of rendering legible text in both Chinese and English characters directly inside generated video frames.
Frequently asked questions
About this listing
This entry was compiled from publicly available data including Wan (Wan 2.1)'s official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Wan (Wan 2.1) unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Wan (Wan 2.1) directly →
Spotted something out of date? Suggest an update →
Wan (Wan 2.1) in the news
More in Video & Audio
