File size: 1,299 Bytes
36977f3
 
 
 
 
 
 
 
 
 
 
26234fe
 
36977f3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
26234fe
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
---
---
language:
- en
- zh
pipeline_tag: text-generation
tags:
- Awq
- int4
- yi1.5-34B-Chat
- pytorch
license: apache-2.0
license_name: apache-2.0
license_link: LICENSE
---

## About Quantization
我们使用modelscope [swift](https://github.com/modelscope/swift/)仓库进行AWQ 4bit量化. 量化文档可以查看[这里](https://github.com/modelscope/swift/blob/main/docs/source/LLM/LLM%E9%87%8F%E5%8C%96%E6%96%87%E6%A1%A3.md). 量化命令如下:

We use the modelscope [swift](https://github.com/modelscope/swift/) repository to perform AWQ 4bit quantization. Quantization documentation can be found [here](https://github.com/modelscope/swift/blob/main/docs/source_en/LLM/LLM-quantization.md). The quantization command is as follows:

```bash
# Experimental Environment: A100
swift export \
    --quant_bits 4 \
    --model_type yi-1_5-34b-chat \
    --quant_method awq \
    --quant_n_samples 32 \
    --dataset alpaca-zh alpaca-en sharegpt-gpt4-mini \
    --quant_seqlen 4096
```

Inference:
```bash
CUDA_VISIBLE_DEVICES=0 swift infer --model_type yi-1_5-34b-chat-awq-int4
```

SFT:
```bash
CUDA_VISIBLE_DEVICES=0 swift sft --model_type yi-1_5-34b-chat-awq-int4 --dataset leetcode-python-en
```

Original Model:

[YI1.5-34B-Chat](https://modelscope.cn/models/01ai/Yi-1.5-34B-Chat/summary)