Update README.md (#1)
Browse files- Update README.md (986a4644d26e4cb680e70ef956922b5d268a085f)
Co-authored-by: Shansan Gong <[email protected]>
README.md
CHANGED
|
@@ -1,3 +1,27 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: unknown
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: unknown
|
| 3 |
+
base_model:
|
| 4 |
+
- Qwen/Qwen2.5-Coder-7B
|
| 5 |
+
tags:
|
| 6 |
+
- code
|
| 7 |
+
- text-diffusion-model
|
| 8 |
+
- diffusion large language model
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
### DiffuCoder-7B-Base
|
| 12 |
+
|
| 13 |
+
The DiffuCoder-7B-Base model is our foundational masked diffusion LLM for code generation.
|
| 14 |
+
|
| 15 |
+
- Training recipe: Using [DiffuLLaMA](https://github.com/HKUNLP/DiffuLLaMA)'s adaptation approach, trained on a large corpus of code: with Stage 1 65B tokens and Stage 2 65B tokens.
|
| 16 |
+
|
| 17 |
+
- Benchmarks: Strong baseline performance on HumanEval, MBPP and BigCodeBench.
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
#### More details and usage examples:
|
| 21 |
+
|
| 22 |
+
- Paper: [DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation](https://arxiv.org/abs/2506.20639)
|
| 23 |
+
|
| 24 |
+
- GitHub: https://github.com/apple/ml-diffucoder
|
| 25 |
+
|
| 26 |
+
#### Acknowledgement
|
| 27 |
+
To power this HuggingFace model release, we reuse [Dream](https://huggingface.co/Dream-org/Dream-v0-Base-7B)'s modeling architecture and generation utils.
|