"docs/vscode:/vscode.git/clone" did not exist on "1244948885e364b3103aa3b515ce66fa4d9955fa"
auto_awq.md 2.73 KB
Newer Older
1
2
3
4
---
title: AutoAWQ
---
[](){ #auto-awq }
5
6

To create a new 4-bit quantized model, you can leverage [AutoAWQ](https://github.com/casper-hansen/AutoAWQ).
7
Quantization reduces the model's precision from BF16/FP16 to INT4 which effectively reduces the total model memory footprint.
8
9
The main benefits are lower latency and memory usage.

10
You can quantize your own models by installing AutoAWQ or picking one of the [6500+ models on Huggingface](https://huggingface.co/models?search=awq).
11

12
```bash
13
pip install autoawq
14
15
```

Reid's avatar
Reid committed
16
After installing AutoAWQ, you are ready to quantize a model. Please refer to the [AutoAWQ documentation](https://casper-hansen.github.io/AutoAWQ/examples/#basic-quantization) for further details. Here is an example of how to quantize `mistralai/Mistral-7B-Instruct-v0.2`:
17

18
??? code
19

20
21
22
    ```python
    from awq import AutoAWQForCausalLM
    from transformers import AutoTokenizer
23

24
25
26
    model_path = 'mistralai/Mistral-7B-Instruct-v0.2'
    quant_path = 'mistral-instruct-v0.2-awq'
    quant_config = { "zero_point": True, "q_group_size": 128, "w_bit": 4, "version": "GEMM" }
27

28
29
30
31
32
    # Load model
    model = AutoAWQForCausalLM.from_pretrained(
        model_path, **{"low_cpu_mem_usage": True, "use_cache": False}
    )
    tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
33

34
35
    # Quantize
    model.quantize(tokenizer, quant_config=quant_config)
36

37
38
39
40
41
42
    # Save quantized model
    model.save_quantized(quant_path)
    tokenizer.save_pretrained(quant_path)

    print(f'Model is quantized and saved at "{quant_path}"')
    ```
43
44
45

To run an AWQ model with vLLM, you can use [TheBloke/Llama-2-7b-Chat-AWQ](https://huggingface.co/TheBloke/Llama-2-7b-Chat-AWQ) with the following command:

46
```bash
Reid's avatar
Reid committed
47
48
49
python examples/offline_inference/llm_engine_example.py \
    --model TheBloke/Llama-2-7b-Chat-AWQ \
    --quantization awq
50
51
52
53
```

AWQ models are also supported directly through the LLM entrypoint:

54
??? code
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79

    ```python
    from vllm import LLM, SamplingParams

    # Sample prompts.
    prompts = [
        "Hello, my name is",
        "The president of the United States is",
        "The capital of France is",
        "The future of AI is",
    ]
    # Create a sampling params object.
    sampling_params = SamplingParams(temperature=0.8, top_p=0.95)

    # Create an LLM.
    llm = LLM(model="TheBloke/Llama-2-7b-Chat-AWQ", quantization="AWQ")
    # Generate texts from the prompts. The output is a list of RequestOutput objects
    # that contain the prompt, generated text, and other information.
    outputs = llm.generate(prompts, sampling_params)
    # Print the outputs.
    for output in outputs:
        prompt = output.prompt
        generated_text = output.outputs[0].text
        print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
    ```