alpayariyak
commited on
Commit
•
8c069a4
1
Parent(s):
7f01910
Update README.md
Browse files
README.md
CHANGED
@@ -6,52 +6,41 @@ colorTo: blue
|
|
6 |
sdk: static
|
7 |
pinned: false
|
8 |
---
|
|
|
9 |
<div align="center">
|
10 |
<h1>OpenChat: Advancing Open-source Language Models with Mixed-Quality Data</h1>
|
11 |
</div>
|
12 |
-
<div align="center">
|
13 |
-
<img src="https://github.com/imoneoi/openchat/raw/master/assets/logo_new.png" style="width: 65%">
|
14 |
-
</div>
|
15 |
|
16 |
-
<p align="center">
|
17 |
-
<
|
18 |
-
<a href="https://github.com/
|
19 |
-
<a href="https://
|
20 |
-
<a href="https://
|
21 |
-
<
|
|
|
|
|
|
|
22 |
<strong>🔥 First 7B model that Achieves Comparable Results with ChatGPT (March)! 🔥</strong>
|
23 |
<br>
|
24 |
<strong>🤖 #1 Open-source model on MT-bench scoring 7.81, outperforming 70B models 🤖</strong>
|
25 |
<br>
|
26 |
</p>
|
27 |
|
28 |
-
<div align="center">
|
29 |
-
<img src="https://github.com/imoneoi/openchat/raw/master/assets/openchat.png" style="width:
|
30 |
-
<img src="https://github.com/imoneoi/openchat/raw/master/assets/openchat_grok.png" style="width:
|
31 |
</div>
|
32 |
|
33 |
- OpenChat is an innovative library of **open-source language models**, fine-tuned with [**C-RLFT**](https://arxiv.org/pdf/2309.11235.pdf) - a strategy inspired by offline reinforcement learning.
|
34 |
- Our models learn from mixed-quality data without preference labels, delivering exceptional performance on par with `ChatGPT`, even with a `7B` model which can be run on a **consumer GPU (e.g. RTX 3090)**.
|
35 |
- Despite our simple approach, we are committed to developing a high-performance, commercially viable, open-source large language model, and we continue to make significant strides toward this vision.
|
36 |
|
37 |
-
[![DOI](https://zenodo.org/badge/645397533.svg)](https://zenodo.org/badge/latestdoi/645397533)
|
38 |
-
|
39 |
# ✨ News
|
40 |
|
41 |
- [2023/11/01] We released the [OpenChat-3.5-7B](https://huggingface.co/openchat/openchat_3.5) model, surpassing ChatGPT on various benchmarks 🔥.
|
42 |
|
43 |
- [2023/09/21] We released our paper [OpenChat: Advancing Open-source Language Models with Mixed-Quality Data](https://arxiv.org/pdf/2309.11235.pdf).
|
44 |
|
45 |
-
- [2023/09/03] We released the [OpenChat V3.2 SUPER]([#models](https://huggingface.co/openchat/openchat_v3.2_super)) model.
|
46 |
-
|
47 |
-
- [2023/08/04] We have launched an [Online Demo](https://openchat.team) featuring the latest version, OpenChat 3.2.
|
48 |
-
|
49 |
-
- [2023/07/30] We are thrilled to introduce the [OpenChat V3 model series](#models), based on Llama 2, and now available for free for commercial use!
|
50 |
-
|
51 |
-
- [2023/07/07] We released the [OpenChat V2 model series](#legacy-models).
|
52 |
-
|
53 |
-
- [2023/07/01] We released the [OpenChat V1 model series](#legacy-models).
|
54 |
-
|
55 |
# 🏷️ Benchmarks
|
56 |
|
57 |
| Model | # Params | Average | MT-Bench | AGIEval | BBH MC | TruthfulQA | MMLU | HumanEval | BBH CoT | GSM8K |
|
|
|
6 |
sdk: static
|
7 |
pinned: false
|
8 |
---
|
9 |
+
|
10 |
<div align="center">
|
11 |
<h1>OpenChat: Advancing Open-source Language Models with Mixed-Quality Data</h1>
|
12 |
</div>
|
|
|
|
|
|
|
13 |
|
14 |
+
<p align="center" style="margin-bottom: 0px;"> <!-- Reduce bottom margin of the paragraph -->
|
15 |
+
<img src="https://github.com/imoneoi/openchat/raw/master/assets/logo_new.png" style="width: 65%">
|
16 |
+
<a href="https://openchat.team"><img src="https://github.com/alpayariyak/openchat/blob/master/logo_new-removebg-preview.png?raw=true" alt="OpenChat Logo" style="width:30px; vertical-align: middle; display: inline-block;"/><span style="display: inline-block; vertical-align: middle; font-size: 20px;">Online Demo of OpenChat 3.5</span></a> |
|
17 |
+
<a href="https://github.com/imoneoi/openchat"><img src="https://camo.githubusercontent.com/4133dc1cd4511d4a292b84ce10e52e4ed92569fb2a8165381c9c47be5edc2796/68747470733a2f2f6564656e742e6769746875622e696f2f537570657254696e7949636f6e732f696d616765732f706e672f6769746875622e706e67" alt="GitHub Logo" style="width:30px; vertical-align: middle; display: inline-block;"/><span style="display: inline-block; vertical-align: middle; font-size: 20px;">GitHub</span></a> |
|
18 |
+
<a href="https://arxiv.org/pdf/2309.11235.pdf"><img src="https://github.com/alpayariyak/openchat/blob/master/arxiv-logomark-small-square-border.png?raw=true" alt="ArXiv Logo" style="width:30px; vertical-align: middle; display: inline-block;"/><span style="display: inline-block; vertical-align: middle; font-size: 20px;">Paper</span></a> |
|
19 |
+
<a href="https://discord.gg/pQjnXvNKHY"><img src="https://cloud.githubusercontent.com/assets/6291467/26705903/96c2d66e-477c-11e7-9f4e-f3c0efe96c9a.png" alt="Discord Logo" style="width:30px; vertical-align: middle; display: inline-block;"/><span style="display: inline-block; vertical-align: middle; font-size: 20px;">Discord</span></a>
|
20 |
+
</p>
|
21 |
+
<!-- Removed <br> tags to reduce space -->
|
22 |
+
<p align="center" style="margin-top: 0px;"> <!-- Reduce top margin of this paragraph -->
|
23 |
<strong>🔥 First 7B model that Achieves Comparable Results with ChatGPT (March)! 🔥</strong>
|
24 |
<br>
|
25 |
<strong>🤖 #1 Open-source model on MT-bench scoring 7.81, outperforming 70B models 🤖</strong>
|
26 |
<br>
|
27 |
</p>
|
28 |
|
29 |
+
<div align="center" style="display: flex; justify-content: center; align-items: center;">
|
30 |
+
<img src="https://github.com/imoneoi/openchat/raw/master/assets/openchat.png" style="width: 35%; margin-right: 2%;">
|
31 |
+
<img src="https://github.com/imoneoi/openchat/raw/master/assets/openchat_grok.png" style="width: 37%;">
|
32 |
</div>
|
33 |
|
34 |
- OpenChat is an innovative library of **open-source language models**, fine-tuned with [**C-RLFT**](https://arxiv.org/pdf/2309.11235.pdf) - a strategy inspired by offline reinforcement learning.
|
35 |
- Our models learn from mixed-quality data without preference labels, delivering exceptional performance on par with `ChatGPT`, even with a `7B` model which can be run on a **consumer GPU (e.g. RTX 3090)**.
|
36 |
- Despite our simple approach, we are committed to developing a high-performance, commercially viable, open-source large language model, and we continue to make significant strides toward this vision.
|
37 |
|
|
|
|
|
38 |
# ✨ News
|
39 |
|
40 |
- [2023/11/01] We released the [OpenChat-3.5-7B](https://huggingface.co/openchat/openchat_3.5) model, surpassing ChatGPT on various benchmarks 🔥.
|
41 |
|
42 |
- [2023/09/21] We released our paper [OpenChat: Advancing Open-source Language Models with Mixed-Quality Data](https://arxiv.org/pdf/2309.11235.pdf).
|
43 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
# 🏷️ Benchmarks
|
45 |
|
46 |
| Model | # Params | Average | MT-Bench | AGIEval | BBH MC | TruthfulQA | MMLU | HumanEval | BBH CoT | GSM8K |
|