hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4

This repository is a community-driven quantized version of the original model mistralai/Mixtral-8x7B-Instruct-v0.1 which is the BF16 half-precision official version released by Mistral AI.

Model Information

The Mixtral-8x7B Large Language Model (LLM) is a pretrained generative Sparse Mixture of Experts. The Mixtral-8x7B outperforms Llama 2 70B on most benchmarks they tested.

For full details of this model please read our release blog post.

This repository contains mistralai/Mixtral-8x7B-Instruct-v0.1 quantized using AutoAWQ from FP16 down to INT4 using the GEMM kernels performing zero-point quantization with a group size of 128.

Model Usage

In order to run the inference with Mixtral 8x7B Instruct AWQ in INT4, around 24 GiB of VRAM are needed only for loading the model checkpoint, excluding the KV cache and/or the CUDA graphs, meaning that there should be a bit over that VRAM available.

In order to use the current quantized model, support is offered for different solutions as transformers, autoawq, or text-generation-inference.

🤗 Transformers

In order to run the inference with Mixtral 8x7B Instruct AWQ in INT4, you need to install the following packages:

pip install -q --upgrade transformers autoawq accelerate

To run the inference on top of Mixtral 8x7B Instruct AWQ in INT4 precision, the AWQ model can be instantiated as any other causal language modeling model via AutoModelForCausalLM and run the inference normally.

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, AwqConfig

model_id = "hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4"

quantization_config = AwqConfig(
    bits=4,
    fuse_max_seq_len=512, # Note: Update this as per your use-case
    do_fuse=True,
)

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
  model_id,
  torch_dtype=torch.float16,
  low_cpu_mem_usage=True,
  device_map="auto",
  quantization_config=quantization_config
)

prompt = [
  {"role": "system", "content": "You are a helpful assistant, that responds as a pirate."},
  {"role": "user", "content": "What's Deep Learning?"},
]
inputs = tokenizer.apply_chat_template(
  prompt,
  tokenize=True,
  add_generation_prompt=True,
  return_tensors="pt",
  return_dict=True,
).to("cuda")

outputs = model.generate(**inputs, do_sample=True, max_new_tokens=256)
print(tokenizer.batch_decode(outputs[:, inputs['input_ids'].shape[1]:], skip_special_tokens=True)[0])

AutoAWQ

In order to run the inference with Mixtral 8x7B Instruct AWQ in INT4, you need to install the following packages:

pip install -q --upgrade transformers autoawq accelerate

Alternatively, one may want to run that via AutoAWQ even though it's built on top of 🤗 transformers, which is the recommended approach instead as described above.

import torch
from awq import AutoAWQForCausalLM
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoAWQForCausalLM.from_pretrained(
  model_id,
  torch_dtype=torch.float16,
  low_cpu_mem_usage=True,
  device_map="auto",
)

prompt = [
  {"role": "system", "content": "You are a helpful assistant, that responds as a pirate."},
  {"role": "user", "content": "What's Deep Learning?"},
]
inputs = tokenizer.apply_chat_template(
  prompt,
  tokenize=True,
  add_generation_prompt=True,
  return_tensors="pt",
  return_dict=True,
).to("cuda")

outputs = model.generate(**inputs, do_sample=True, max_new_tokens=256)
print(tokenizer.batch_decode(outputs[:, inputs['input_ids'].shape[1]:], skip_special_tokens=True)[0])

The AutoAWQ script has been adapted from AutoAWQ/examples/generate.py.

🤗 Text Generation Inference (TGI)

To run the text-generation-launcher with Mixtral 8x7B Instruct AWQ in INT4 with Marlin kernels for optimized inference speed, you will need to have Docker installed (see installation notes) and the huggingface_hub Python package as you need to login to the Hugging Face Hub.

pip install -q --upgrade huggingface_hub
huggingface-cli login

Then you just need to run the TGI v2.0.3 (or higher) Docker container as follows:

docker run --gpus all --shm-size 1g -ti -p 8080:80 \
  -v hf_cache:/data \
  -e MODEL_ID=hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4 \
  -e QUANTIZE=awq \
  -e HF_TOKEN=$(cat ~/.cache/huggingface/token) \
  -e MAX_INPUT_LENGTH=4000 \
  -e MAX_TOTAL_TOKENS=4096 \
  ghcr.io/huggingface/text-generation-inference:2.0.3

TGI will expose different endpoints, to see all the endpoints available check TGI OpenAPI Specification.

To send request to the deployed TGI endpoint compatible with OpenAI OpenAPI specification i.e. /v1/chat/completions:

curl 0.0.0.0:8080/v1/chat/completions \
  -X POST \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "tgi",
    "messages": [
      {
        "role": "system",
        "content": "You are a helpful assistant."
      },
      {
        "role": "user",
        "content": "What is Deep Learning?"
      }
    ],
    "max_tokens": 128
  }'

Or programatically via the huggingface_hub Python client as follows:

import os
from huggingface_hub import InferenceClient

client = InferenceClient(base_url="http://0.0.0.0:8080", api_key=os.getenv("HF_TOKEN", "-"))

chat_completion = client.chat.completions.create(
  model="hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4",
  messages=[
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "What is Deep Learning?"},
  ],
  max_tokens=128,
)

Alternatively, the OpenAI Python client can also be used (see installation notes) as follows:

import os
from openai import OpenAI

client = OpenAI(base_url="http://0.0.0.0:8080/v1", api_key=os.getenv("OPENAI_API_KEY", "-"))

chat_completion = client.chat.completions.create(
  model="tgi",
  messages=[
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "What is Deep Learning?"},
  ],
  max_tokens=128,
)

vLLM

To run vLLM with Mixtral 8x7B Instruct AWQ in INT4, you will need to have Docker installed (see installation notes) and run the latest vLLM Docker container as follows:

docker run --runtime nvidia --gpus all --ipc=host -p 8000:8000 \
  -v hf_cache:/root/.cache/huggingface \
  vllm/vllm-openai:latest \
  --model hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4 \
  --max-model-len 4096

To send request to the deployed vLLM endpoint compatible with OpenAI OpenAPI specification i.e. /v1/chat/completions:

curl 0.0.0.0:8000/v1/chat/completions \
  -X POST \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4",
    "messages": [
      {
        "role": "system",
        "content": "You are a helpful assistant."
      },
      {
        "role": "user",
        "content": "What is Deep Learning?"
      }
    ],
    "max_tokens": 128
  }'

Or programatically via the openai Python client (see installation notes) as follows:

import os
from openai import OpenAI

client = OpenAI(base_url="http://0.0.0.0:8000/v1", api_key=os.getenv("VLLM_API_KEY", "-"))

chat_completion = client.chat.completions.create(
  model="hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4",
  messages=[
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "What is Deep Learning?"},
  ],
  max_tokens=128,
)

Quantization Reproduction

In order to quantize Mixtral 8x7B Instruct using AutoAWQ, you will need to use an instance with at least enough CPU RAM to fit the whole model in half-precision i.e. ~90GiB, and an NVIDIA GPU with at least 16GiB of VRAM to quantize it.

In order to quantize Mixtral 8x7B Instruct, first install the following packages:

pip install -q --upgrade transformers autoawq accelerate

Then run the following script, adapted from AutoAWQ/examples/quantize.py:

from awq import AutoAWQForCausalLM
from transformers import AutoTokenizer

model_path = "mistralai/Mixtral-8x7B-Instruct-v0.1"
quant_path = "hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4"
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)

# 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}"')