AutoAWQ

要创建新的 4 位量化模型，您可以利用 AutoAWQ。量化将模型的精度从 FP16 降低到 INT4，这有效地将文件大小减少约 70%。主要优点是更低的延迟和内存使用。

您可以通过安装 AutoAWQ 或选择 Huggingface 上的 400 多个模型来量化您自己的模型。

pip install autoawq

安装 AutoAWQ 后，您就可以量化模型了。以下是如何量化 mistralai/Mistral-7B-Instruct-v0.2 的示例

from awq import AutoAWQForCausalLM
from transformers import AutoTokenizer

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" }

# 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)

# Quantize
model.quantize(tokenizer, quant_config=quant_config)

# Save quantized model
model.save_quantized(quant_path)
tokenizer.save_pretrained(quant_path)

print(f'Model is quantized and saved at "{quant_path}"')

要使用 vLLM 运行 AWQ 模型，您可以使用 TheBloke/Llama-2-7b-Chat-AWQ 以及以下命令

python examples/offline_inference/llm_engine_example.py --model TheBloke/Llama-2-7b-Chat-AWQ --quantization awq

LLM 入口点也直接支持 AWQ 模型

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}")