MODEL Alibaba Last updated:

Wan 2.x

Alibaba's Open-Source Video Family 阿里开源视频家族（通义万相）

Alibaba's open-source video family — Wan 2.2 → 2.5 → 2.6 → 2.7 — leads the open VBench leaderboard. Wan 2.7 (early 2026) added a "Thinking Mode" that plans complex multi-shot videos before generating, similar to Veo 3 Ingredients. Apache 2.0 throughout. 阿里的开源视频模型家族，从 Wan 2.2 一路走到 2.5 / 2.6 / 2.7， 目前稳居开源阵营 VBench 榜首。2026 年初推出的 Wan 2.7 加入 "Thinking Mode"，在生成前先规划多镜头叙事，思路对标 Veo 3 的 Ingredients。整条线一律 Apache 2.0。

Try Qwen API Docs ↗
Cost
Free
Open weights — self-host
How are Intelligence, Speed & Cost bucketed?
Intelligence and Speed buckets are percentile ranks on Artificial Analysis. Cost buckets are fixed dollar thresholds keyed off output-token price ($/M out).
Intelligence
  • Top 1%≤ 1%
  • Top 5%≤ 5%
  • Top 10%≤ 10%
  • Good≤ 25%
  • Medium≤ 50%
  • Below avg> 50%
Speed
  • Top 1%≥ 345 tok/s
  • Top 5%≥ 237 tok/s
  • Top 10%≥ 196 tok/s
  • Good≥ 146 tok/s
  • Medium≥ 90 tok/s
  • Slow< 90 tok/s
Cost
  • Freeopen weights · self-host
  • Low< $1 / M out
  • Moderate$1–5 / M out
  • High≥ $5 / M out
Official ↗ GitHub ↗

Why it matters

Currently the strongest fully-open video model on VBench. Pushed HunyuanVideo and other open competitors to keep up. Made open video generation a real procurement option for media / agencies.

当下 VBench 上最强的全开源视频模型，把腾讯 HunyuanVideo 等开源对手逼着追赶；也让"开源视频生成"从玩票变成媒体 / 代理商真正可采买的选项。

Core Capabilities

Generative
Produces images, video, audio, or other media.
Multimodal
Combines text, vision, and audio in one model.
Vision
Understands images, scenes, and visual context.

Context Window

Context window not disclosed.

Availability

API
Not available
Product / App
Not available
Open Source
Released
Enterprise
—

Pricing Model

Free / self-host
Open weights — pay only for compute.
Self-host

Capability / Performance

Where this model sits relative to the middle 60% of models in the tree. All scores are 0–10 (higher is better).

Lower 20% Upper 80% This model
Lower 20% 20th percentile — 20% of models score below this This model Where the current model lands Upper 80% 80th percentile — only 20% of models score above this Percentile boundaries are computed across every model in the tree that reports the underlying benchmark for each capability.

What it feels like

  • Vision-language model from Alibaba — see the linked sources below for benchmark and review coverage
  • Vision and multimodal tasks are the typical fit per the published model card

Best use cases

  • Vision tasks (charts, documents, images) per the model card
  • See the model spec and sources block for benchmarked use cases

Tools to try

Qwen Chat Hugging Face Ollama DashScope

Not ideal for

  • Tasks far outside the modalities listed in this model's spec
  • Workflows where a more recent successor in the same family scores higher