---
license: apache-2.0
base_model: Qwen/Qwen3.5-2B
---

# VTR-2B

Visual Token Routing (VTR) on Qwen3.5-2B. A small router, trained jointly with the model, decides which visual tokens
enter the language model. VTR matches the full-resolution model on OSWorld, CUA-World and video game action prediction
with 2.5 to 5.8× fewer visual tokens.

- Paper: [arXiv:2611.06854](https://arxiv.org/abs/2611.06854)
- Base model: [Qwen/Qwen3.5-2B](https://huggingface.co/Qwen/Qwen3.5-2B)
- Router: three-layer MLP with 23.3M parameters, about 1% of the language model
- Training: 2,000 steps of joint training
- Every 10th image is kept in full

## Usage

The router runs through the patched vLLM server in the [VTR repository](https://github.com/pranjal2041/vtr).

```bash
git clone https://github.com/pranjal2041/vtr
cd vtr/inference
uv sync
uv run vtr patch
uv run vtr serve l3lab/VTR-2B --port 8000
```

The server is OpenAI-compatible. See `inference/README.md` for the fast mode and other options.