# Tencent releases WeVisOne-VL-4B vision-language model on Hugging Face

> The new 4‑billion‑parameter model handles object detection, segmentation and image matting from a single prompt.

Oossa · 2026-10-23 · https://oossa.com/en/tencent-releases-wevisone-vl-4b-vision-language-model-on-hugging-face

Tencent’s WeChat Vision team added the WeVisOne-VL-4B model to Hugging Face. It builds on Qwen3‑VL and can do open‑vocabulary object detection, counting, OCR, referring segmentation and alpha‑matte extraction in one step. The model includes a lightweight 9.6 M‑parameter pixel decoder and uses special coordinate tokens to return bounding boxes and masks. The release is aimed at research use only; it is not meant for autonomous high‑impact decisions.

## The facts

- Model size: 4 billion parameters, plus a 9.6 M‑parameter pixel decoder
- Release date: 2026‑10‑22 (published on Hugging Face)

## Why it matters

Researchers can now experiment with a single model for many vision‑language tasks without stitching together separate systems.

## Sources & references

1. [New model on Hugging Face: tencent/WeVisOne-VL-4B (image-text-to-text)](https://huggingface.co/tencent/WeVisOne-VL-4B) – Tencent Hunyuan, 2026-10-22

Last updated: 2026-10-23