UoomiNumPy-deepseek-1.5B-educational-Project-Quant Recovery
Authors/Creators
Description
UoomiNumPy-deepseek-1.5B Educational Project
Quickly dissect the technical foundations of the DeepSeek-R1-1.5B large language model using UoomiNumPy, then run it directly on CPU for instant testing without any frameworks.
👨💻 Author
**uxiaosu** (TensorLinx / Shanghai Lingxi Matrix Technology Co., Ltd.)
📧 Contact: contact@tensorlinx.com
🔗 Project: https://github.com/uxiaosu/UoomiNumPy-deepseek-1.5B-Educational-Project
---
📝 Academic Citation
If you use this project in academic research, please cite:
```bibtex
@software{uoominum_deepseek_2025,
author = {uxiaosu},
title = {UoomiNumPy-deepseek-1.5B-Educational-Project: A Pure NumPy Implementation of DeepSeek Language Model},
year = {2025},
organization = {TensorLinx / Shanghai Lingxi Matrix Technology Co., Ltd.},
url = {https://github.com/uxiaosu/UoomiNumPy-deepseek-1.5B-Educational-Project}
}
```
Files
deepseek_numpy_weights.zip
Files
(3.1 GB)
| Name | Size | Download all |
|---|---|---|
|
md5:ec46b79b2661be7ea46408eaeab92e6f
|
3.1 GB | Preview Download |
Additional details
Software
- Programming language
- Python
- Development Status
- Unsupported