This is the Python client package for Javelin.

For more information about Javelin, see https://getjavelin.com

Javelin Documentation: https://docs.getjavelin.io

For local development, Please change version = "RELEASE_VERSION" with any semantic version example : version = "v0.1.10" in pyproject.toml

Make sure that the file pyproject.toml reverted before commit back to main

  pip install javelin-sdk

# Create virtual environment
python -m venv venv

# Activate virtual environment
venv\Scripts\activate

# Install dependencies
pip install poetry
poetry install

# Create virtual environment
python -m venv venv

# Activate virtual environment
source venv/bin/activate

# Install dependencies
pip install poetry
poetry install

# Uninstall any existing version
pip uninstall javelin-sdk -y

# Build the package
poetry build

# Install the newly built package
pip install dist/javelin_sdk-<version>-py3-none-any.whl

Javelin provides universal endpoints that allow you to use a consistent interface across different LLM providers. Here are the main patterns:

• Basic Azure OpenAI integration
• Universal endpoint implementation
• OpenAI-compatible interface

• Basic Bedrock integration
• Universal endpoint implementation
• OpenAI-compatible interface

• Basic Gemini integration
• Universal endpoint implementation
• OpenAI-compatible interface

• CrewAI integration
• LangGraph integration

• Asynchronous example
• Synchronous example
• Drop-in replacement example

• Document processing
• RAG implementation

For more detailed examples and integration patterns, check out:

• Azure OpenAI Integration
• AWS Bedrock Integration
• Supported Language Models