bedrock()

Create an AWS Bedrock language model

Factory function to create an AWS Bedrock model instance using the Converse API.

Requires optional dependency: pip install ai-query[bedrock]

Signature

def bedrock(
    model_id: str,
    *,
    region: str | None = None,
    credentials: dict | Callable | None = None,
) -> LanguageModel

Parameters

model_id · str · required

The Bedrock model ID (e.g., “anthropic.claude-3-5-sonnet-20241022-v2:0”).

region · str

AWS region. Falls back to AWS_REGION or AWS_DEFAULT_REGION environment variable.

credentials · dict | Callable

AWS credentials. Can be:

• None: Use default AWS credential chain
• dict: {"access_key": "...", "secret_key": "...", "session_token": "..."}
• Callable: A credentials provider function

Returns

A LanguageModel instance configured for AWS Bedrock.

Environment

Uses the standard AWS credential chain. Optionally set:

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

Available Models

Model ID	Description
anthropic.claude-3-5-sonnet-20241022-v2:0	Claude 3.5 Sonnet v2
anthropic.claude-3-haiku-20240307-v1:0	Claude 3 Haiku
meta.llama3-1-70b-instruct-v1:0	Llama 3.1 70B
amazon.nova-pro-v1:0	Amazon Nova Pro

Example

from ai_query import generate_text
from ai_query.providers.bedrock import bedrock

result = await generate_text(
    model=bedrock("anthropic.claude-3-5-sonnet-20241022-v2:0"),
    prompt="Hello, world!"
)

With Credentials

result = await generate_text(
    model=bedrock(
        "anthropic.claude-3-5-sonnet-20241022-v2:0",
        region="us-west-2",
        credentials={
            "access_key": "AKIA...",
            "secret_key": "..."
        }
    ),
    prompt="Hello!"
)