anonymitaet commited on
Commit
44e52a8
β€’
1 Parent(s): 6bfcc89

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +92 -0
README.md ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ ---
4
+ <div align="center">
5
+
6
+ <picture>
7
+ <img src="https://raw.githubusercontent.com/01-ai/Yi/main/assets/img/Yi_logo_icon_light.svg" width="150px">
8
+ </picture>
9
+
10
+ </div>
11
+
12
+ <p align="center">
13
+ <a href="https://github.com/01-ai">πŸ™ GitHub</a> β€’
14
+ <a href="https://discord.gg/hYUwWddeAu">πŸ‘Ύ Discord</a> β€’
15
+ <a href="https://twitter.com/01ai_yi">🐀 Twitter</a> β€’
16
+ <a href="https://github.com/01-ai/Yi-1.5/issues/2">πŸ’¬ WeChat</a>
17
+ <br/>
18
+ <a href="https://arxiv.org/abs/2403.04652">πŸ“ Paper</a> β€’
19
+ <a href="https://github.com/01-ai/Yi/tree/main?tab=readme-ov-file#faq">πŸ™Œ FAQ</a> β€’
20
+ <a href="https://github.com/01-ai/Yi/tree/main?tab=readme-ov-file#learning-hub">πŸ“— Learning Hub</a>
21
+ </p>
22
+
23
+ # Intro
24
+
25
+ Yi-1.5 is an upgraded version of Yi. It is continuously pre-trained on Yi with a high-quality corpus of 500B tokens and fine-tuned on 3M diverse fine-tuning samples.
26
+
27
+ Compared with Yi, Yi-1.5 delivers stronger performance in coding, math, reasoning, and instruction-following capability, while still maintaining excellent capabilities in language understanding, commonsense reasoning, and reading comprehension.
28
+
29
+ <div align="center">
30
+
31
+ Model | Context Length | Pre-trained Tokens
32
+ | :------------: | :------------: | :------------: |
33
+ | Yi-1.5 | 4K, 16K, 32K | 3.6T
34
+
35
+ </div>
36
+
37
+ # Models
38
+
39
+ - Chat models
40
+
41
+ <div align="center">
42
+
43
+ | Name | Download |
44
+ | --------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
45
+ | Yi-1.5-34B-Chat | β€’ [πŸ€— Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β€’ [πŸ€– ModelScope](https://www.modelscope.cn/organization/01ai) |
46
+ | Yi-1.5-34B-Chat-16K | β€’ [πŸ€— Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β€’ [πŸ€– ModelScope](https://www.modelscope.cn/organization/01ai) |
47
+ | Yi-1.5-9B-Chat | β€’ [πŸ€— Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β€’ [πŸ€– ModelScope](https://www.modelscope.cn/organization/01ai) |
48
+ | Yi-1.5-9B-Chat-16K | β€’ [πŸ€— Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β€’ [πŸ€– ModelScope](https://www.modelscope.cn/organization/01ai) |
49
+ | Yi-1.5-6B-Chat | β€’ [πŸ€— Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β€’ [πŸ€– ModelScope](https://www.modelscope.cn/organization/01ai) |
50
+
51
+ </div>
52
+
53
+ - Base models
54
+
55
+ <div align="center">
56
+
57
+ | Name | Download |
58
+ | ---------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
59
+ | Yi-1.5-34B | β€’ [πŸ€— Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β€’ [πŸ€– ModelScope](https://www.modelscope.cn/organization/01ai) |
60
+ | Yi-1.5-34B-32K | β€’ [πŸ€— Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β€’ [πŸ€– ModelScope](https://www.modelscope.cn/organization/01ai) |
61
+ | Yi-1.5-9B | β€’ [πŸ€— Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β€’ [πŸ€– ModelScope](https://www.modelscope.cn/organization/01ai) |
62
+ | Yi-1.5-9B-32K | β€’ [πŸ€— Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β€’ [πŸ€– ModelScope](https://www.modelscope.cn/organization/01ai) |
63
+ | Yi-1.5-6B | β€’ [πŸ€— Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β€’ [πŸ€– ModelScope](https://www.modelscope.cn/organization/01ai) |
64
+
65
+ </div>
66
+
67
+ # Benchmarks
68
+
69
+ - Chat models
70
+
71
+ Yi-1.5-34B-Chat is on par with or excels beyond larger models in most benchmarks.
72
+
73
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/KcsJ9Oc1VnEmfCDEJc5cd.png)
74
+
75
+ Yi-1.5-9B-Chat is the top performer among similarly sized open-source models.
76
+
77
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/xf6pLg5jqRCwjlh6m3t6_.png)
78
+
79
+ - Base models
80
+
81
+ Yi-1.5-34B is on par with or excels beyond larger models in some benchmarks.
82
+
83
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/BwU7QM-03dZvZzwdIE1xY.png)
84
+
85
+ Yi-1.5-9B is the top performer among similarly sized open-source models.
86
+
87
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/y-EYSYPT-3aWLJ0x8R94F.png)
88
+
89
+ # Quick Start
90
+
91
+ For getting up and running with Yi-1.5 models quickly, see [README](https://github.com/01-ai/Yi-1.5).
92
+