Tencent/WeVisOne-VL
WeVisOne-VL is a powerful VLM by the WeChat Vision Team at Tencent, unifying object-level and pixel-level perception from points and boxes to masks and alpha mattes.
Older snapshot: check the original source for current information.
From the publisher
README & documentation
Source previewRead the project’s overview, installation instructions and usage examples. The original README is the source of truth.
English | 简体中文 Headline numbers for WeVisOne-VL-4B against the strongest open baseline in each benchmark (Qwen3-VL / Qwen3.5 up to 9B, Rex-Omni, LocateAnything, and task-specific segmentation / matting models). F1 for detection, referring and OCR is averaged over IoU thresholds 0.5:0.95 ("Mean"). Greedy decoding, ≤ 1,960 visual tokens per image. Object Detection (F1 Mean) Referring Object Detection (F1 Mean / Rejection Rate %) Object Counting (Exact-match Accuracy %) Object Pointing OCR — Text Grounding & Recognition (F1 Mean) Referring Segmentation (cIoU; ReasonSeg uses gIoU) Promptable Matting (MAD ×10³ ↓) See the technical report for all splits and metrics, and evaluation/README.md to reproduce the public benchmarks. 1. Data Engine. A knowledge-graph-driven engine discovers concepts, localizes them with complementary experts verified by a VLM, and extends the verified boxes to masks and alpha mattes. Over 100M images are annotated this way. 2. Unified Task Formulation. Every task is written as structured text,…
Read the full README ↗ · Preview checked 2026-10-27T13:26:23.185Z
Inside the original README — Document outline
- WeVisOne-VL
- Unified Visual Perception from Objects to Pixels
- ✨ Highlights
- 📢 News
- 📝 TODO
- 🏆 Performance at a Glance
- 📊 Full Benchmark Results (click each task to expand)
- 🏗️ How It Works
- 🚀 Quick Start
- Installation
- Inference
- Supported Tasks
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the README
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: Other. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
What this repository does
WeVisOne-VL is a powerful VLM by the WeChat Vision Team at Tencent, unifying object-level and pixel-level perception from points and boxes to masks and alpha mattes.
Repository facts
- Owner
- Tencent
- Primary language
- Jupyter Notebook
- Stars
- 419
- Forks
- 37
- Open issues + pull requests
- 6
- License
- Other
- Archived
- No
- Default branch
- main
- Created
- 2026-10-22T08:58:16.000Z
- Last push
- 2026-10-25T10:17:41.000Z
Topics and intended use
No topics were included in the latest source metadata.
Review the README for scope, installation, examples and limitations. We do not run repository code or certify it.
Evaluate before installing
Review licensing and dependencies, inspect recent commits and unresolved issues, and test in an isolated environment before production use. Stars and forks alone cannot answer these questions.
Source and freshness
Source: GitHub. Metadata observed 2026-10-29T12:15:50.303Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.