RichardErkhov
commited on
Commit
β’
51d87f9
1
Parent(s):
2dc91c7
uploaded readme
Browse files
README.md
ADDED
@@ -0,0 +1,112 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Quantization made by Richard Erkhov.
|
2 |
+
|
3 |
+
[Github](https://github.com/RichardErkhov)
|
4 |
+
|
5 |
+
[Discord](https://discord.gg/pvy7H8DZMG)
|
6 |
+
|
7 |
+
[Request more models](https://github.com/RichardErkhov/quant_request)
|
8 |
+
|
9 |
+
|
10 |
+
Yi-1.5-9B-32K - bnb 4bits
|
11 |
+
- Model creator: https://huggingface.co/01-ai/
|
12 |
+
- Original model: https://huggingface.co/01-ai/Yi-1.5-9B-32K/
|
13 |
+
|
14 |
+
|
15 |
+
|
16 |
+
|
17 |
+
Original model description:
|
18 |
+
---
|
19 |
+
license: apache-2.0
|
20 |
+
---
|
21 |
+
<div align="center">
|
22 |
+
|
23 |
+
<picture>
|
24 |
+
<img src="https://raw.githubusercontent.com/01-ai/Yi/main/assets/img/Yi_logo_icon_light.svg" width="150px">
|
25 |
+
</picture>
|
26 |
+
|
27 |
+
</div>
|
28 |
+
|
29 |
+
<p align="center">
|
30 |
+
<a href="https://github.com/01-ai">π GitHub</a> β’
|
31 |
+
<a href="https://discord.gg/hYUwWddeAu">πΎ Discord</a> β’
|
32 |
+
<a href="https://twitter.com/01ai_yi">π€ Twitter</a> β’
|
33 |
+
<a href="https://github.com/01-ai/Yi-1.5/issues/2">π¬ WeChat</a>
|
34 |
+
<br/>
|
35 |
+
<a href="https://arxiv.org/abs/2403.04652">π Paper</a> β’
|
36 |
+
<a href="https://01-ai.github.io/">πͺ Tech Blog</a> β’
|
37 |
+
<a href="https://github.com/01-ai/Yi/tree/main?tab=readme-ov-file#faq">π FAQ</a> β’
|
38 |
+
<a href="https://github.com/01-ai/Yi/tree/main?tab=readme-ov-file#learning-hub">π Learning Hub</a>
|
39 |
+
</p>
|
40 |
+
|
41 |
+
# Intro
|
42 |
+
|
43 |
+
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.
|
44 |
+
|
45 |
+
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.
|
46 |
+
|
47 |
+
<div align="center">
|
48 |
+
|
49 |
+
Model | Context Length | Pre-trained Tokens
|
50 |
+
| :------------: | :------------: | :------------: |
|
51 |
+
| Yi-1.5 | 4K, 16K, 32K | 3.6T
|
52 |
+
|
53 |
+
</div>
|
54 |
+
|
55 |
+
# Models
|
56 |
+
|
57 |
+
- Chat models
|
58 |
+
|
59 |
+
<div align="center">
|
60 |
+
|
61 |
+
| Name | Download |
|
62 |
+
| --------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
63 |
+
| Yi-1.5-34B-Chat | β’ [π€ Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β’ [π€ ModelScope](https://www.modelscope.cn/organization/01ai) β’ [π£ wisemodel](https://wisemodel.cn/organization/01.AI)|
|
64 |
+
| Yi-1.5-34B-Chat-16K | β’ [π€ Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β’ [π€ ModelScope](https://www.modelscope.cn/organization/01ai) β’ [π£ wisemodel](https://wisemodel.cn/organization/01.AI) |
|
65 |
+
| Yi-1.5-9B-Chat | β’ [π€ Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β’ [π€ ModelScope](https://www.modelscope.cn/organization/01ai) β’ [π£ wisemodel](https://wisemodel.cn/organization/01.AI) |
|
66 |
+
| Yi-1.5-9B-Chat-16K | β’ [π€ Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β’ [π€ ModelScope](https://www.modelscope.cn/organization/01ai) β’ [π£ wisemodel](https://wisemodel.cn/organization/01.AI) |
|
67 |
+
| Yi-1.5-6B-Chat | β’ [π€ Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β’ [π€ ModelScope](https://www.modelscope.cn/organization/01ai) β’ [π£ wisemodel](https://wisemodel.cn/organization/01.AI) |
|
68 |
+
|
69 |
+
</div>
|
70 |
+
|
71 |
+
- Base models
|
72 |
+
|
73 |
+
<div align="center">
|
74 |
+
|
75 |
+
| Name | Download |
|
76 |
+
| ---------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
77 |
+
| Yi-1.5-34B | β’ [π€ Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β’ [π€ ModelScope](https://www.modelscope.cn/organization/01ai) β’ [π£ wisemodel](https://wisemodel.cn/organization/01.AI) |
|
78 |
+
| Yi-1.5-34B-32K | β’ [π€ Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β’ [π€ ModelScope](https://www.modelscope.cn/organization/01ai) β’ [π£ wisemodel](https://wisemodel.cn/organization/01.AI) |
|
79 |
+
| Yi-1.5-9B | β’ [π€ Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β’ [π€ ModelScope](https://www.modelscope.cn/organization/01ai) β’ [π£ wisemodel](https://wisemodel.cn/organization/01.AI) |
|
80 |
+
| Yi-1.5-9B-32K | β’ [π€ Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β’ [π€ ModelScope](https://www.modelscope.cn/organization/01ai) β’ [π£ wisemodel](https://wisemodel.cn/organization/01.AI) |
|
81 |
+
| Yi-1.5-6B | β’ [π€ Hugging Face](https://huggingface.co/collections/01-ai/yi-15-2024-05-663f3ecab5f815a3eaca7ca8) β’ [π€ ModelScope](https://www.modelscope.cn/organization/01ai) β’ [π£ wisemodel](https://wisemodel.cn/organization/01.AI) |
|
82 |
+
|
83 |
+
</div>
|
84 |
+
|
85 |
+
# Benchmarks
|
86 |
+
|
87 |
+
- Chat models
|
88 |
+
|
89 |
+
Yi-1.5-34B-Chat is on par with or excels beyond larger models in most benchmarks.
|
90 |
+
|
91 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/KcsJ9Oc1VnEmfCDEJc5cd.png)
|
92 |
+
|
93 |
+
Yi-1.5-9B-Chat is the top performer among similarly sized open-source models.
|
94 |
+
|
95 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/xf6pLg5jqRCwjlh6m3t6_.png)
|
96 |
+
|
97 |
+
- Base models
|
98 |
+
|
99 |
+
Yi-1.5-34B is on par with or excels beyond larger models in some benchmarks.
|
100 |
+
|
101 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/BwU7QM-03dZvZzwdIE1xY.png)
|
102 |
+
|
103 |
+
Yi-1.5-9B is the top performer among similarly sized open-source models.
|
104 |
+
|
105 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/y-EYSYPT-3aWLJ0x8R94F.png)
|
106 |
+
|
107 |
+
# Quick Start
|
108 |
+
|
109 |
+
For getting up and running with Yi-1.5 models quickly, see [README](https://github.com/01-ai/Yi-1.5).
|
110 |
+
|
111 |
+
|
112 |
+
|