Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- vision-language model
|
4 |
+
- llama
|
5 |
+
- yi
|
6 |
+
- generation
|
7 |
+
datasets:
|
8 |
+
- YanweiLi/Mini-Gemini-Instruction
|
9 |
+
---
|
10 |
+
|
11 |
+
# Mini-Gemini-34B Model Card
|
12 |
+
<a href='https://github.com/dvlab-research/MiniGemini'><img src='https://img.shields.io/badge/Project-Code-violet'></a>
|
13 |
+
<a href='https://mini-gemini.github.io/'><img src='https://img.shields.io/badge/Project-Page-Green'></a>
|
14 |
+
<a href='https://arxiv.org/pdf/2403.18814.pdf'><img src='https://img.shields.io/badge/Paper-Arxiv-red'></a>
|
15 |
+
|
16 |
+
## Model details
|
17 |
+
Mini-Gemini supports a series of dense and MoE Large Language Models (LLMs) from 2B to 34B with HD image understanding, reasoning, and generation simultaneously.
|
18 |
+
You can also try our other Mini-Gemini series models:
|
19 |
+
|
20 |
+
Normal resolution setting: [Mini-Gemini-2B](https://huggingface.co/YanweiLi/Mini-Gemini-2B), [Mini-Gemini-7B](https://huggingface.co/YanweiLi/Mini-Gemini-7B), [Mini-Gemini-13B](https://huggingface.co/YanweiLi/Mini-Gemini-13B), [Mini-Gemini-8x7B](https://huggingface.co/YanweiLi/Mini-Gemini-8x7B)
|
21 |
+
|
22 |
+
High resolution setting: [Mini-Gemini-7B-HD](https://huggingface.co/YanweiLi/Mini-Gemini-7B-HD), [Mini-Gemini-13B](https://huggingface.co/YanweiLi/Mini-Gemini-13B-HD), [Mini-Gemini-8x7B-HD](https://huggingface.co/YanweiLi/Mini-Gemini-8x7B-HD), [Mini-Gemini-34B-HD](https://huggingface.co/YanweiLi/Mini-Gemini-34B-HD)
|
23 |
+
|
24 |
+
**Model type:**
|
25 |
+
Mini-Gemini is an open-source chatbot trained by fine-tuning Nous-Hermes-2-Yi-34B on GPT-generated multimodal instruction-following data.
|
26 |
+
|
27 |
+
Mini-Gemini empowers existing frameworks to support HD image understanding, reasoning, and generation simultaneously.
|
28 |
+
|
29 |
+
**Model version:**
|
30 |
+
Mini-Gemini with LLM Nous-Hermes-2-Yi-34B
|
31 |
+
|
32 |
+
**Model date:**
|
33 |
+
Mini-Gemini-34B was trained on 03/2024.
|
34 |
+
|
35 |
+
## License
|
36 |
+
Nous-Hermes-2-Yi-34B is licensed under the apache-2.0 License,
|
37 |
+
|
38 |
+
**Where to send questions or comments about the model:**
|
39 |
+
https://github.com/dvlab-research/MiniGemini/issues
|
40 |
+
|
41 |
+
## Intended use
|
42 |
+
**Primary intended uses:**
|
43 |
+
The primary use of Mini-Gemini is research on large multimodal models and chatbots.
|
44 |
+
|
45 |
+
**Primary intended users:**
|
46 |
+
The primary intended users of the model are researchers and hobbyists in computer vision, natural language processing, machine learning, and artificial intelligence.
|
47 |
+
|
48 |
+
## Training data
|
49 |
+
This model is trained based on [Mini-Gemini-Instruction](https://huggingface.co/datasets/YanweiLi/Mini-Gemini-Instruction) dataset, please to the [Github](https://github.com/dvlab-research/MiniGemini) for more detail.
|