Amazon Bedrock

Amazon Bedrock can be configured easily using the elsai-model package.

Install

pip install --extra-index-url https://core-packages.elsai.ai/root/elsai-model/ elsai-model

Prerequisites

• AWS account with Bedrock access enabled
• Model access enabled for your chosen model in the AWS console
• AWS credentials configured (via environment variables, ~/.aws/credentials, or IAM role)

Basic usage

import os
from elsai import Agent
from elsai_model.bedrock import BedrockConnector

model = BedrockConnector(
    aws_access_key=os.environ.get("AWS_ACCESS_KEY_ID"),
    aws_secret_key=os.environ.get("AWS_SECRET_ACCESS_KEY"),
    aws_region="us-east-1",
    model_id="anthropic.claude-3-5-sonnet-20241022-v2:0"
)

agent = Agent(model=model)
result = agent("Hello!")

Specify a model

You can initialize BedrockConnector with any supported Bedrock model ID:

from elsai_model.bedrock import BedrockConnector

model = BedrockConnector(model_id="us.amazon.nova-pro-v1:0")

Configuration

import os
from elsai_model.bedrock import BedrockConnector

model = BedrockConnector(
    aws_access_key=os.environ.get("AWS_ACCESS_KEY_ID"),
    aws_secret_key=os.environ.get("AWS_SECRET_ACCESS_KEY"),
    aws_region="us-east-1",
    model_id="anthropic.claude-3-5-sonnet-20241022-v2:0",
    temperature=0.3,
    max_tokens=4096,
)

Parameters

Parameter	Type	Default	Description
aws_access_key	str | None	None	AWS access key
aws_secret_key	str | None	None	AWS secret key
aws_region	str | None	None	AWS region (e.g. us-east-1)
model_id	str | None	None	Bedrock model ID
max_tokens	int	500	Max tokens to generate
temperature	float	0.1	Sampling temperature

Popular model IDs

Model	Model ID
Claude 3.5 Sonnet v2	anthropic.claude-3-5-sonnet-20241022-v2:0
Amazon Nova Pro	us.amazon.nova-pro-v1:0
Amazon Nova Lite	us.amazon.nova-lite-v1:0
Llama 3.1 70B	us.meta.llama3-1-70b-instruct-v1:0
Mistral Large	us.mistral.mistral-large-2402-v1:0