Summary
- Alibaba's Wan video generation model family announced via its official account that Wan3.0 has entered public beta.
- The release highlights native generation of 30-second videos and "reality-grade" rendering as core features.
- It introduces Omni-Reference, expanding reference inputs from text, image, audio, and video to documents, spreadsheets, slides, and web pages.
Alibaba's Wan Enters Public Beta with 3.0
Alibaba's Wan video generation model family has opened public beta access to its next version, Wan3.0. According to a post on its official account, the release was introduced with the tagline "Simple Input. Smart Creation." and made immediately accessible to general users at the beta stage. The post was published on August 6, 2026 (UTC).
Native 30-Second Generation and Rendering Quality
The first item on the announced feature list is native generation of 30-second videos. This was presented as generating a full 30 seconds in one pass rather than stitching together multiple clips, though the announcement did not include implementation details or specifications such as resolution or frame rate. The second item, "Reality-Grade Rendering," was used to describe near-photorealistic rendering quality, but no supporting benchmark figures or comparison points were disclosed.
Omni-Reference: Expanded Reference Inputs
The most notable item is Omni-Reference. Beyond the existing reference types of text, image, audio, and video, the model reportedly accepts documents, spreadsheets, slides, and web pages as reference inputs. This appears to point toward directly incorporating tabular data, presentations, and web documents as source material for video production, though the original post was cut off mid-sentence, so it is unconfirmed whether the listed input types end there.
| Item | Wan3.0 Announcement Details |
|---|---|
| Release form | Public beta |
| Video length | Native 30 seconds |
| Rendering | Reality-Grade Rendering |
| Reference inputs | Text, image, audio, video, documents, spreadsheets, slides, web pages |
| Disclosed metrics | None |
What Remains Unconfirmed
The announcement made no mention of whether model weights would be released, supported languages, access channels, or commercial usage terms. Regardless of whatever distribution approach the Wan family has used in the past, the currently available material provides no basis for assuming the same will apply to Wan3.0. Since this information is based on a single post, specifications may become clearer once a technical report or official documentation is released.
The competitive landscape for video generation models and trends among recently released multimodal models can be followed further in metallab.ai's video generation coverage.
Points to Watch
The axes of competition in video generation are shifting toward length, realism, and controllability. The three features Wan3.0 emphasized correspond respectively to these three axes. In particular, the structure of accepting documents and spreadsheets as references can be seen as designed for production workflows—such as advertising, reports, and educational content—where source material already exists. However, actual quality and generation stability can only be judged once results accumulate from beta users.
Comments