Commit
·
e0106af
1
Parent(s):
f613971
Update README.md
Browse files
README.md
CHANGED
|
@@ -68,13 +68,14 @@ pinned: false
|
|
| 68 |
<br>🏆 <u>#1</u> Open-source model on MT-bench scoring 7.81, outperforming <u>70B</u> models 🏆
|
| 69 |
</span>
|
| 70 |
</a>
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
<div align="center" style="display: flex; justify-content: center; align-items: center; "'>
|
| 75 |
<img src="https://github.com/imoneoi/openchat/raw/master/assets/openchat.png" style="width:45%; margin-right: 2%;">
|
| 76 |
<img src="https://github.com/imoneoi/openchat/raw/master/assets/openchat_grok.png" style="width: 47%;">
|
| 77 |
</div>
|
|
|
|
|
|
|
|
|
|
|
|
|
| 78 |
|
| 79 |
- 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.
|
| 80 |
- 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)**.
|
|
|
|
| 68 |
<br>🏆 <u>#1</u> Open-source model on MT-bench scoring 7.81, outperforming <u>70B</u> models 🏆
|
| 69 |
</span>
|
| 70 |
</a>
|
| 71 |
+
<div align="center" style="display: flex; justify-content: center; align-items: center; "'>
|
|
|
|
|
|
|
|
|
|
| 72 |
<img src="https://github.com/imoneoi/openchat/raw/master/assets/openchat.png" style="width:45%; margin-right: 2%;">
|
| 73 |
<img src="https://github.com/imoneoi/openchat/raw/master/assets/openchat_grok.png" style="width: 47%;">
|
| 74 |
</div>
|
| 75 |
+
</p>
|
| 76 |
+
|
| 77 |
+
|
| 78 |
+
|
| 79 |
|
| 80 |
- 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.
|
| 81 |
- 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)**.
|