Bedrock in Orbit

Amazon Bedrock provides access to foundation models from AWS, including Anthropic, Nova, and more. Orbit supports Bedrock as a backend for agent orchestration and generative AI workflows.

Key Features

• Multiple model providers: Access a range of LLMs and embedding models via Bedrock.
• Secure integration: Use AWS IAM roles and credentials for secure access.
• Scalable inference: Run high-throughput, low-latency inference for production workloads.

Example: Bedrock API Call

import boto3

client = boto3.client("bedrock-runtime")
response = client.invoke_model(
    modelId="anthropic.claude-v2",
    body={"prompt": "Hello, Claude!", "max_tokens": 100}
)
print(response["body"].read())

Best Practices

• Use SSO or credential vendor for AWS authentication.
• Monitor usage and costs in the AWS console.
• Combine Bedrock with LangChain or Orbit Agents for advanced workflows.

See the Orbit documentation for more Bedrock integration patterns and security guidance.