AI Agent StoreTencent · Download documented
WeVisOne-VL · 2B and 4B
Find, count or outline objects in a still image, locate text, or separate a subject from its background. Choose the 2B or 4B download and use Tencent’s inference package.
First reviewed 2026-10-27 · Checked 2026-11-03. The launch date has not been confirmed.
We still need comparable independent tests. You can compare the confirmed features and limits below, but we cannot yet give this version an Agent Fit or Practical Value rating. Try it on your hardware and tasks before relying on it.
What you can use today
You download the model files and supply the hardware and software to run them. Capacity is the amount of information a request can contain; settings and available memory can limit what fits.
Exact download ID: tencent/WeVisOne-VL-2B / tencent/WeVisOne-VL-4B. Returns structured locations and image masks. The mask decoder is a required component. This release targets one image-perception turn at a time, without an explicit thinking trace; its pixel decoder rejects video grids.
Intended for visual-perception research, not identifying individuals or autonomous high-impact decisions. Boundaries can be unreliable with transparency, occlusion or low-resolution images. The repository labels Apache 2.0; third-party components retain their own terms.
A usable request or maximum output allowance is not established by the card. Do not substitute the parent Qwen3-VL model’s limits.
You supply compatible hardware and software. No hosted token rate or independent comparison is established.
Provider release and specifications ↗
Check availability and plans ↗
More setup and licensing details ↗
Test it before switching
Give each model the same examples, tools and definition of success. Record correct outcomes, response time, retries and total cost. Include the time and hardware needed to run it.
Compare caching and long-context costs ↗