Vllm rollingbatch deploy mixtral 8x7b

Run this notebook online:

vllm rollingbatch Mixtral-8x7B deployment guide

In this tutorial, you will use LMI container from DLC to SageMaker and run inference with it.

Please make sure the following permission granted before running the notebook:

• SageMaker access

Step 1: Let's bump up SageMaker and import stuff

%pip install sagemaker --upgrade  --quiet

import os
import sagemaker
from sagemaker.djl_inference.model import DJLModel

role = sagemaker.get_execution_role()  # execution role for the endpoint
session = sagemaker.session.Session()  # sagemaker session for interacting with different AWS APIs

Step 2: Start building SageMaker endpoint

In this step, we will build SageMaker endpoint from scratch

Getting the container image URI (optional)

Check out available images: Large Model Inference available DLC

# Choose a specific version of LMI image directly:
# image_uri = "763104351884.dkr.ecr.us-west-2.amazonaws.com/djl-inference:0.28.0-lmi10.0.0-cu124"

Create SageMaker model

Here we are using LMI PySDK to create the model.

Checkout more configuration options.

model_id = "mistralai/Mixtral-8x7B-v0.1" # model will be download form Huggingface hub
hf_token = os.getenv("HF_TOKEN", "hf_XXXXXXXXXXX")    # use your HF_TOKEN to access this model

env = {
    "TENSOR_PARALLEL_DEGREE": "8",         # requires 8 GPUs, set to "max" to use all GPUs on the instance
    "HF_TOKEN": hf_token,
    "OPTION_ROLLING_BATCH": "vllm",        # enable rollingbatch with vllm
}

model = DJLModel(
            model_id=model_id,
            env=env,
            role=role)

Create SageMaker endpoint

You need to specify the instance to use and endpoint names

instance_type = "ml.g5.48xlarge"
endpoint_name = sagemaker.utils.name_from_base("lmi-model")

predictor = model.deploy(initial_instance_count=1,
             instance_type=instance_type,
             endpoint_name=endpoint_name,
             # container_startup_health_check_timeout=3600,
            )

Step 3: Run inference

predictor.predict(
    {"inputs": "The future of Artificial Intelligence is", "parameters": {"max_new_tokens":128, "do_sample":"true"}}
)

Clean up the environment

session.delete_endpoint(endpoint_name)
session.delete_endpoint_config(endpoint_name)
model.delete_model()