mistralai

/

Mixtral-8x7B-Instruct-v0.1

like 3.8k

Trying to deploy on SageMaker with

import json
import sagemaker
import boto3
from sagemaker.huggingface import HuggingFaceModel, get_huggingface_llm_image_uri

try:
    role = sagemaker.get_execution_role()
except ValueError:
    iam = boto3.client('iam')
    role = iam.get_role(RoleName='sagemaker_execution_role')['Role']['Arn']

# Hub Model configuration. https://huggingface.co/models
model_id = 'mistralai/Mixtral-8x7B-Instruct-v0.1'
hub = {
    'HF_MODEL_ID': model_id,
    'SM_NUM_GPUS': json.dumps(8)
}



# create Hugging Face Model Class
huggingface_model = HuggingFaceModel(
    image_uri=get_huggingface_llm_image_uri("huggingface"),
    transformers_version="4.36.0",
    env=hub,
    role=role, 
    name=f"HF-{model_id}".replace('/','-').replace('.','-')
)

# deploy model to SageMaker Inference
predictor = huggingface_model.deploy(
    initial_instance_count=1,
    instance_type="ml.g5.48xlarge",
    container_startup_health_check_timeout=300,
  )

But get the following error

File "/opt/conda/lib/python3.9/site-packages/text_generation_server/server.py", line 159, in serve_inner model = get_model( File "/opt/conda/lib/python3.9/site-packages/text_generation_server/models/__init__.py", line 291, in get_model raise ValueError("sharded is not supported for AutoModel")

ValueError: sharded is not supported for AutoModel

Any ideas on how to fix?