Open source · 5 November 2026
Available nowTencent releases WeVisOne-VL vision-language model in 2B and 4B sizes
Tencent’s WeChat Vision team released WeVisOne-VL, an Apache 2.0 vision‑language model available in 2 billion‑parameter and 4 billion‑parameter versions. The model handles detection, OCR, counting, segmentation and alpha matting in a single architecture, using a 9.6 million‑parameter pixel decoder.
Why it matters
WeVisOne-VL’s unified capabilities let you prototype a single visual‑AI service covering detection, OCR, segmentation and matting, so prioritize building multi‑task pipelines now.
For product managers
For your product
What does this mean for your product?
Try it on your product More from this issueAll AI news from 5 Nov →
Get this every morning, written for your product.
Deltaday reads the whole AI news cycle and tells you what changed, whether it's real yet, and what to do about it.