llama-index-llms-bedrock-converse

What it is and what it does

This package bridges LlamaIndex and AWS Bedrock's Converse API, allowing you to build LLM applications that call foundation models (Claude, Command, Mistral) through Bedrock instead of directly. It wraps boto3 calls and integrates with LlamaIndex's LLM interface, so you can use Bedrock models anywhere LlamaIndex expects an LLM—in RAG pipelines, agents, or standalone completions.

The package supports the full Bedrock Converse feature set: streaming completions and chat, function calling with tool integration, prompt caching to reduce costs on repeated context, and async operations. You authenticate via AWS profiles, access keys, or session tokens, and can optionally route requests through Application Inference Profiles for provisioned capacity.

Use it for:

• Build RAG pipelines using Claude or other Bedrock models within LlamaIndex without vendor lock-in to direct API calls.
• Stream LLM responses in real-time chat applications by calling Bedrock's streaming endpoints through LlamaIndex.
• Enable agents to call external functions via Bedrock's native function-calling support integrated with LlamaIndex tools.
• Reduce inference costs on repeated queries by leveraging Bedrock's prompt caching with strategic cache points.
• Use provisioned Bedrock capacity via Application Inference Profiles for predictable, scaled workloads.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Integrates AWS Bedrock's Converse API with LlamaIndex, enabling LLM applications to call foundation models like Claude through Bedrock with support for streaming, function calling, and prompt caching.

Yes. This is a straightforward, actively maintained integration with low install friction and no known vulnerabilities. Install it if you're already using LlamaIndex and want to run Bedrock models; skip it if you're calling Bedrock directly or using a different LLM framework.

Install

llama-index-llms-bedrock-converse on PyPI

pip install llama-index-llms-bedrock-converse

uv add llama-index-llms-bedrock-converse

poetry add llama-index-llms-bedrock-converse

Installing llama-index-llms-bedrock-converse

Before you install

Installs with low friction—only two runtime dependencies (boto3 and llama-index-core). Actively maintained with a recent release.

License in practice

MIT license permits commercial and private use with minimal restrictions.

Quickstart

from llama_index.llms.bedrock_converse import BedrockConverse

llm = BedrockConverse(
    model="anthropic.claude-3-haiku-20240307-v1:0",
    profile_name="your_aws_profile"
)
resp = llm.complete("Paul Graham is ")
print(resp)

Requires AWS credentials (profile, access keys, or session token) and Bedrock model access configured in your AWS account.

Verify before relying

• Whether all Bedrock foundation models are supported or only specific ones.
• Performance characteristics and latency when streaming large responses.
• Cost implications of using Bedrock vs. direct API calls.

Package facts

Evidence: llama_index_llms_bedrock_converse-0.14.18-py3-none-any.whl

Tags

More Artificial Intelligence packages

Further reading