Skip to content

AWS Bedrock

AWS Bedrock provides access to multiple foundation models (Claude, Llama, Titan, Mistral, Cohere) through AWS infrastructure with enterprise security and compliance.

Stable

Quick Facts

Property Value
Models Claude, Llama, Titan, Mistral, Cohere
Security AWS IAM, VPC, encryption
Compliance SOC, HIPAA, FedRAMP
Streaming Supported
Functions Supported
Vision Supported (Claude)

Setup

Prerequisites

  1. AWS Account with Bedrock access
  2. IAM credentials with Bedrock permissions
  3. Enable models in Bedrock console

Configure

export AWS_ACCESS_KEY_ID="..."
export AWS_SECRET_ACCESS_KEY="..."
export AWS_REGION="us-east-1"

from sentimatrix.config import SentimatrixConfig, LLMConfig

config = SentimatrixConfig(
    llm=LLMConfig(
        provider="bedrock",
        model="anthropic.claude-3-5-sonnet-20241022-v2:0",
        aws_region="us-east-1"
    )
)

llm:
  provider: bedrock
  model: anthropic.claude-3-5-sonnet-20241022-v2:0
  aws_region: us-east-1

Available Models

Provider Model ID Best For
Anthropic anthropic.claude-3-5-sonnet-20241022-v2:0 General, reasoning
Anthropic anthropic.claude-3-haiku-20240307-v1:0 Fast, cost-effective
Meta meta.llama3-70b-instruct-v1:0 Open-source quality
Amazon amazon.titan-text-premier-v1:0 AWS-native
Mistral mistral.mistral-large-2407-v1:0 European compliance
Cohere cohere.command-r-plus-v1:0 RAG, embeddings

Usage Examples

Basic Usage

import asyncio
from sentimatrix import Sentimatrix
from sentimatrix.config import SentimatrixConfig, LLMConfig

config = SentimatrixConfig(
    llm=LLMConfig(
        provider="bedrock",
        model="anthropic.claude-3-5-sonnet-20241022-v2:0"
    )
)

async def main():
    async with Sentimatrix(config) as sm:
        insights = await sm.generate_insights(reviews)
        print(insights)

asyncio.run(main())

With IAM Role

config = SentimatrixConfig(
    llm=LLMConfig(
        provider="bedrock",
        model="anthropic.claude-3-5-sonnet-20241022-v2:0",
        aws_region="us-east-1",
        # Uses default credential chain (IAM role, env vars, etc.)
    )
)

Configuration Options

LLMConfig(
    provider="bedrock",
    model="anthropic.claude-3-5-sonnet-20241022-v2:0",

    # AWS settings
    aws_region="us-east-1",
    aws_access_key_id="...",  # Optional if using IAM role
    aws_secret_access_key="...",

    # Generation settings
    temperature=0.7,
    max_tokens=4096,

    # Reliability
    timeout=60,
    max_retries=3,
)

Best Practices

  1. Use IAM Roles in Production

    • Avoid hardcoded credentials
    • Use EC2 instance roles or EKS service accounts

  2. Choose Region Carefully

    • Not all models available in all regions
    • Consider latency and data residency

  3. Enable Model Access

    • Models must be enabled in Bedrock console
    • Some require approval