# Wan: what AI models say (September 2026)

Source: Memetik Index, https://www.memetik.ai/vendors/wan
Website: https://wan.video

Wan was named in 7 of 50 AI answers across 1 category. Best position: AI video generation, 14% answer share, ranked #12, named first in 0%.

## What Wan is

Wan is a set of open-source models for generating and editing video. It is built for industrial applications and academic research, including work that needs video creation, transformation or animation from text, images, reference material and voice.

The models support text-to-video and image-to-video generation. Wan’s VACE model can generate video from references, edit existing video and apply masked video-to-video edits. Its animation model handles character animation and character replacement. An audio-driven model generates cinematic video from voice.

(From Wan's own website, read 2026-09-02.)

## By category

| Category | Rank | Answer share | Named first | Models |
|---|---|---|---|---|
| AI video generation (https://www.memetik.ai/index/ai-video-generation) | #12 of 16 | 14% | 0% | Claude Opus 5, Claude, Claude Fable 5, Gemini 3.5 Flash |

## What the models said

> "Alibaba's Wan 2.7 is the strongest model you can run on your own hardware, appealing to developers who want local control rather than cloud APIs."
> — Claude, AI video generation

> "Open-Source Options Wan 2.2 and LTX-2.3 now produce output rivaling Sora and Veo on motion realism."
> — Claude Opus 5, AI video generation