Initial commit
Browse files
README.md
CHANGED
|
@@ -1,10 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |

|
| 2 |
|
| 3 |
<p align="center">
|
| 4 |
🏠 <a href="https://codegeex.cn" target="_blank">Homepage</a>|🛠 Tools <a href="https://marketplace.visualstudio.com/items?itemName=aminer.codegeex" target="_blank">VS Code</a>, <a href="https://plugins.jetbrains.com/plugin/20587-codegeex" target="_blank">Jetbrains</a>|🤗 <a href="https://huggingface.co/THUDM/codegeex2-6b" target="_blank">HF Repo</a>|📄 <a href="https://arxiv.org/abs/2303.17568" target="_blank">Paper</a>|👋 Join our <a href="https://wj.qq.com/s2/11274205/a15b/"target="_blank">Wechat</a>
|
| 5 |
</p>
|
| 6 |
|
| 7 |
-
# CodeGeeX2: 更强大的多语言代码生成模型
|
|
|
|
| 8 |
|
| 9 |
CodeGeeX2 是多语言代码生成模型 [CodeGeeX](https://github.com/THUDM/CodeGeeX) ([KDD’23](https://arxiv.org/abs/2303.17568)) 的第二代模型。CodeGeeX2 基于 [ChatGLM2](https://github.com/THUDM/ChatGLM2-6B) 架构加入代码预训练实现,得益于 ChatGLM2 的更优性能,CodeGeeX2 在多项指标上取得性能提升(+107% > CodeGeeX;仅60亿参数即超过150亿参数的 StarCoder-15B 近10%),更多特性包括:
|
| 10 |
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- zh
|
| 4 |
+
- en
|
| 5 |
+
tags:
|
| 6 |
+
- codegeex
|
| 7 |
+
- glm
|
| 8 |
+
- chatglm
|
| 9 |
+
- thudm
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |

|
| 13 |
|
| 14 |
<p align="center">
|
| 15 |
🏠 <a href="https://codegeex.cn" target="_blank">Homepage</a>|🛠 Tools <a href="https://marketplace.visualstudio.com/items?itemName=aminer.codegeex" target="_blank">VS Code</a>, <a href="https://plugins.jetbrains.com/plugin/20587-codegeex" target="_blank">Jetbrains</a>|🤗 <a href="https://huggingface.co/THUDM/codegeex2-6b" target="_blank">HF Repo</a>|📄 <a href="https://arxiv.org/abs/2303.17568" target="_blank">Paper</a>|👋 Join our <a href="https://wj.qq.com/s2/11274205/a15b/"target="_blank">Wechat</a>
|
| 16 |
</p>
|
| 17 |
|
| 18 |
+
# CodeGeeX2: 更强大的多语言代码生成模型
|
| 19 |
+
# A More Powerful Multilingual Code Generation Model
|
| 20 |
|
| 21 |
CodeGeeX2 是多语言代码生成模型 [CodeGeeX](https://github.com/THUDM/CodeGeeX) ([KDD’23](https://arxiv.org/abs/2303.17568)) 的第二代模型。CodeGeeX2 基于 [ChatGLM2](https://github.com/THUDM/ChatGLM2-6B) 架构加入代码预训练实现,得益于 ChatGLM2 的更优性能,CodeGeeX2 在多项指标上取得性能提升(+107% > CodeGeeX;仅60亿参数即超过150亿参数的 StarCoder-15B 近10%),更多特性包括:
|
| 22 |
|