---
library_name: transformers
license: apache-2.0
base_model: Qwen/Qwen3-4B
tags:
- web-scraping
- information-extraction
- qwen
- qwen3
- co-scraper
pipeline_tag: text-generation
---

# Qwen-HTML 4B

Qwen-HTML is a lightweight HTML-specialized model for **Co-Scraper**, a query-aware web data extraction framework. It is initialized from [Qwen/Qwen3-4B](https://huggingface.co/Qwen/Qwen3-4B) and trained for two complementary tasks:

1. **Node identification**: select the DOM nodes most likely to contain query-relevant records and attributes.
2. **Scraper synthesis**: generate reusable Python extraction code over the pruned HTML.

The model is designed to reduce the cost and failure rate of web information extraction by working with compact, query-aware DOM views instead of full raw pages.

## Model description

- **Base model:** Qwen3-4B
- **Architecture:** `Qwen3ForCausalLM`
- **Precision:** bfloat16
- **Context length:** 131,072 tokens

## Intended use

Deploy the model with [vLLM](https://docs.vllm.ai/) as an OpenAI-compatible service:

```bash
pip install vllm

vllm serve Ambition0303/Qwen-HTML \
  --served-model-name qwen-html \
  --dtype bfloat16 \
  --tensor-parallel-size 1 \
  --max-model-len 131072 \
  --port 8000
```

Then use the endpoint with the [Co-Scraper](https://github.com/Ambition-0303/Co-Scraper) inference pipeline:

```bash
export OPENAI_BASE_URL="http://localhost:8000/v1"
export OPENAI_API_KEY="not-required"

coscraper \
  "Extract the camera model and price." \
  schema.json \
  seed_1.html seed_2.html seed_3.html \
  --model qwen-html
```

Set `--tensor-parallel-size` to the number of GPUs used for tensor parallelism. If GPU memory is insufficient for the full 131,072-token context, reduce `--max-model-len` according to your deployment budget.

## Evaluation

Results from the Co-Scraper paper. Higher is better except Unex.f.

| Dataset | Precision | Recall | F1 | Cor<sub>p</sub> | Cor<sub>f</sub> | Unex.<sub>f</sub> |
|---|---:|---:|---:|---:|---:|---:|
| SWDE | 95.93 | 96.24 | 96.00 | 85.87 | 95.51 | 3.19 |
| LiveWeb-IE | 92.29 | 92.68 | 92.40 | 82.00 | 91.28 | 6.08 |

Metric definitions:

- **F1** ： field-level extraction quality.
- **Cor<sub>p</sub>** ： page-level perfect extraction rate.
- **Cor<sub>f</sub>** ： field-level cross-page reuse success rate.
- **Unex.<sub>f</sub>** ： fraction of fields with unexecutable or invalid output.

## Limitations

Generated scrapers should still be reviewed before being executed on untrusted pages. Model performance can vary with HTML quality, site structure, and query clarity.

## Citation

If you use this model, please cite the Co-Scraper work:

```bibtex
@misc{coscraper2026,
  title={Co-Scraper: Query-Aware DOM Pruning and Reusable Scraper Synthesis for Lightweight Web Data Extraction},
  author={Wang, Shoupeng and Qiu, Jiantao and Zhang, Wuyang and He, Conghui},
  year={2026},
  url={https://arxiv.org/abs/2606.14821}