LLaMA 3.1 405B용 추론 서버
모델을 테스트하기 위해 기본적인 Python 예제부터 시작해 보겠습니다:
from vllm import LLM
# Load LLaMA 3.1 405B on 8 GPUs
llm = LLM("neuralmagic/Meta-Llama-3.1-405B-Instruct-FP8-dynamic", tensor_parallel_size=8)
print(llm.generate("What is the difference between fp8 quantization and int8 quantization?"))
Python 스크립트를 실행할 수 있습니다. 처음 실행하는 경우 모델이 다운로드되어 GPU에 로드될 때까지 기다린 다음 다음과 같은 응답을 받게 됩니다:
FP8 (Floating-Point 8) and INT8 (Integer 8) are both quantization techniques used to reduce the precision of numerical values in deep learning models, but they differ in their representation and behavior.
**INT8 Quantization**
INT8 quantization represents numbers as 8-bit integers, which can take on values between -128 and 127 (or 0 and 255 for unsigned integers). This means that the precision of the numbers is limited to 8 bits, and any values outside this range are clipped or saturated.
INT8 quantization is a simple and widely used technique, especially for integer-based architectures like ARM and x86. Most deep learning frameworks, including TensorFlow and PyTorch, support INT8 quantization.
**FP8 Quantization**
FP8 quantization, on the other hand, represents numbers as 8-bit floating-point numbers, with 1 sign bit, 2 exponent bits, and 5 mantissa bits. This allows for a much larger dynamic range than INT8, with values that can be as small as 2^-14 or as large as 2^15.
FP8 quantization is a more recent development, and its main advantage is that it can provide better accuracy than INT8 quantization, especially for models that require a large dynamic range, such as those with batch normalization or depthwise separable convolutions. FP8 is also more suitable for models that are sensitive to quantization noise, like those with recurrent neural networks (RNNs) or long short-term memory (LSTM) networks.
**Key differences**
Here are the key differences between FP8 and INT8 quantization:
1. **Dynamic range**: FP8 has a much larger dynamic range than INT8, which means it can represent a wider range of values.
2. **Precision**: FP8 has a lower precision than INT8, with 5 mantissa bits compared to 8 bits for INT8.
3. **Behavior**: FP8 is more suitable for models that require a large dynamic range, while INT8 is better suited for models with smaller weights and activations.
4. **Hardware support**: INT8 is widely supported by most hardware platforms, while FP8 is still an emerging standard, with limited hardware support.
In summary, FP8 quantization offers better accuracy and a larger dynamic range than INT8 quantization, but it requires more sophisticated hardware support and may not be suitable for all models or applications.
LLaMA 3.1 405B 모델은 Neural Magic에 의해 이미 vLLM용 fp8로 양자화되었으므로 양자화를 다시 수행할 필요가 없습니다. HuggingFace 허브에서 양자화된 모델을 로드하기만 하면 됩니다.
tensor_parallel_size 매개변수는 머신에서 보유하고 있는 GPU 수에 따라 설정됩니다.
하지만 이 간단한 Python 스크립트는 적절한 프로덕션 서버가 아닙니다. 이제 많은 요청을 병렬로 처리하고 처리량을 최대화하기 위해 추론 서버를 시작하겠습니다:
python -m vllm.entrypoints.openai.api_server \
--model neuralmagic/Meta-Llama-3.1-405B-Instruct-FP8-dynamic \
--tensor-parallel-size 8
모델이 로드되면 두 번째 터미널을 시작하고 몇 가지 요청을 할 수 있습니다:
curl http://localhost:8000/v1/completions \
-H "Content-Type: application/json" \
-d '{
"model": "neuralmagic/Meta-Llama-3.1-405B-Instruct-FP8-dynamic",
"prompt": "Who are you?"
}'
I'm an artificial intelligence model known as Llama. Llama stands for "Large Language Model Meta AI."
