update readme
Browse files
README.md
CHANGED
|
@@ -1,3 +1,28 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
pipeline_tag: text-generation
|
| 4 |
+
datasets:
|
| 5 |
+
- liuhaotian/LLaVA-Pretrain
|
| 6 |
+
- liuhaotian/LLaVA-Instruct-150K
|
| 7 |
---
|
| 8 |
+
# 😈 Imp
|
| 9 |
+
|
| 10 |
+
## Introduction
|
| 11 |
+
In order to fit MLC, we employed 4-bit quantization on the basis of [`Imp-v1.5-3B-196`](https://huggingface.co/MILVLG/Imp-v1.5-3B-196-q4f16_1-MLC) to get `Imp-v1.5-3B-196-q4f16_1-MLC`
|
| 12 |
+
|
| 13 |
+
## License
|
| 14 |
+
This project is licensed under the Apache License 2.0 - see the [LICENSE](https://www.apache.org/licenses/LICENSE-2.0) file for details.
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
## Citation
|
| 18 |
+
|
| 19 |
+
If you use our model or refer our work in your studies, please cite:
|
| 20 |
+
|
| 21 |
+
```bibtex
|
| 22 |
+
@article{imp2024,
|
| 23 |
+
title={Imp: Highly Capable Large Multimodal Models for Mobile Devices},
|
| 24 |
+
author={Shao, Zhenwei and Yu, Zhou and Yu, Jun and Ouyang, Xuecheng and Lihao, Zheng and Zhenbiao, Gai and Mingyang, Wang and Jiajun, Ding},
|
| 25 |
+
journal={arXiv preprint arXiv:2405.12107},
|
| 26 |
+
year={2024}
|
| 27 |
+
}
|
| 28 |
+
```
|