metadata

base_model:
  - Qwen/Qwen2.5-Coder-7B-Instruct
datasets:
  - luzimu/WebGen-Bench
language:
  - en
library_name: transformers
license: mit
metrics:
  - accuracy
pipeline_tag: text-generation
tags:
  - code-generation

WebGen-LM

WebGen-LM is trained using the Bolt.diy trajectories generated from a subset of the training set of WebGen-Bench (🤗 luzimu/WebGen-Bench). It has been introduced in the paper WebGen-Bench: Evaluating LLMs on Generating Interactive and Functional Websites from Scratch.

The training data and code can be found at WebGen-Bench (Github).

The WebGen-LM family of models are as follows:

Models	HF Links
WebGen-LM-7B	🤗 luzimu/WebGen-LM-7B
WebGen-LM-14B	🤗 luzimu/WebGen-LM-14B
WebGen-LM-32B	🤗 luzimu/WebGen-LM-32B

Performance on WebGen-Bench

Sample Usage

You can use this model with the Hugging Face transformers library.

from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline

model_id = "luzimu/WebGen-LM-7B" # This model card refers to WebGen-LM-7B

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")

# Example for website generation
user_prompt = "Generate a simple HTML page with a heading 'Hello, World!' and a paragraph of lorem ipsum text."
messages = [
    {"role": "user", "content": user_prompt}
]

# Apply chat template for instruction-following format
text_input = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)

# Generate output
model_inputs = tokenizer(text_input, return_tensors="pt").to(model.device)
generated_ids = model.generate(model_inputs.input_ids, max_new_tokens=500, do_sample=True, temperature=0.01, top_k=50, top_p=0.95)

# Decode and print the generated code
generated_text = tokenizer.decode(generated_ids[0], skip_special_tokens=True)
print(generated_text)

# Example using Hugging Face pipeline for simpler inference
generator = pipeline("text-generation", model=model, tokenizer=tokenizer)
result = generator(user_prompt, max_new_tokens=500, do_sample=True, temperature=0.01, top_k=50, top_p=0.95)
print(result[0]['generated_text'])

Citation

If you find our project useful, please cite:

@misc{lu2025webgenbenchevaluatingllmsgenerating,
      title={WebGen-Bench: Evaluating LLMs on Generating Interactive and Functional Websites from Scratch}, 
      author={Zimu Lu and Yunqiao Yang and Houxing Ren and Haotian Hou and Han Xiao and Ke Wang and Weikang Shi and Aojun Zhou and Mingjie Zhan and Hongsheng Li},
      year={2025},
      eprint={2505.03733},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2505.03733}, 
}