WAND-Vega
Tencent Cloud launched WAND-Vega, a multimodal application model, with a Real-time Intelligence service and a contact-center voice agent, at its Global Digital Ecosystem Summit on October 30.
The upgraded WAND-Vega is built on mainstream open-source models that Tencent optimized, and it generates and understands images, video and speech for advertising, e-commerce, short dramas and comics. It joins Hunyuan models and more than 100 media processing tools in the WAND audio and video AI brand. Real-time Intelligence adds speech recognition, synthesis and real-time rendering with conversation and understanding agents.
- Date
- Friday 30 October 2026
- Lab
- Tencent Cloud
- Kind
- product
- Access
- closed API
Figures
| Measure | Value | Measured by |
|---|---|---|
| Companies served by WAND | more than 2,000 | company |
| Images generated per day on WAND | more than 4.5 million | company |
| Video generated per day on WAND | 3 million seconds | company |
| Content production cost reduction | about 20% customers overall | company |
| DroidUp Moya first response speedup with EdgeOne Booster | 30% faster interaction jitter cut 4.6 times | company |
Companies call it through existing Tencent Cloud products and standard APIs. The base open-source models are not named.
Sources
This record was checked against its sources on 2 November 2026. How we check