---
language:
- code
library_name: transformers
license: apache-2.0
pipeline_tag: text-generation
tags:
- masked-diffusion
- code-generation
- qwen2
---

## Open Diffusion Large Language Models for Code Generation

This repository contains the weights and custom code for the **fredzzp/open-dcoder-0.5B** model, a masked diffusion model for code generation based on the Qwen2 architecture. 

The model was introduced in the paper [Don't Retrain, Align: Adapting Autoregressive LMs to Diffusion LMs via Representation Alignment](https://huggingface.co/papers/2605.06885).

- **Code:** [pengzhangzhi/Open-dLLM](https://github.com/pengzhangzhi/Open-dLLM)
- **Blog:** [Notion Blog](https://oval-shell-31c.notion.site/Open-Diffusion-Large-Language-Model-25e03bf6136480b7a4ebe3d53be9f68a?pvs=74)

This model uses bidirectional attention and must be used with the custom `diffusion_generate` method.

## How to Use

First, make sure you have the latest `transformers` library installed.

```bash
pip install transformers torch huggingface_hub
```

You can then use the model for generation. Note: You must pass `trust_remote_code=True` to load the custom model architecture.

```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "fredzzp/open-dcoder-0.5B"
device = "cuda" if torch.cuda.is_available() else "cpu"

# trust_remote_code=True is essential
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    trust_remote_code=True
).to(device)

prompt = "def fibonacci(n):"
input_ids = tokenizer(prompt, return_tensors="pt").input_ids.to(device)

# The model will use the generation_config.json from the repo by default
# You can also override parameters here
outputs = model.diffusion_generate(
    inputs=input_ids,
    max_new_tokens=100,
    steps=16,
    temperature=0.8
)

# Decode the output
prompt_len = input_ids.shape[1]
generated_text = tokenizer.decode(outputs.sequences[0][prompt_len:], skip_special_tokens=True)

print("--- Generated Code ---")
print(generated_text)
```

## Citation

```bibtex
@misc{opendllm2025,
  title        = {Open-dLLM: Open Diffusion Large Language Models},
  author       = {Fred Zhangzhi Peng, Shuibai Zhang, Alex Tong, and contributors},
  year         = {2025},
  howpublished = {\url{https://github.com/pengzhangzhi/Open-dLLM}},
  note         = {Blog: \url{https://oval-shell-31c.notion.site/Open-Diffusion-Large-Language-Model-25e03bf6136480b7a4ebe3d53be9f68a?pvs=74}, 
                  Model: \url{https://huggingface.co/fredzzp/open-dcoder-0.5B}}
}
```