llama-find

Llama Find (llama-find) is a sophisticated search and information retrieval toolkit within the LlamaSearch AI ecosystem. It provides agents capable of querying multiple underlying search engines, processing results, and delivering synthesized information through a unified API. It includes support for MLX optimizations.

Key Features

• Agent-Based Search: Utilizes intelligent agents to manage search queries.
• Multi-Engine Support: Integrates with various search backends (defined in search_engines/).
• API Access: Exposes functionality through a comprehensive API (api.py).
• Web Interface (Optional): Includes components for a basic web server (web_server.py).
• MLX Compatibility: Offers potential performance benefits on compatible hardware (mlx_compat.py).
• Caching: Implements caching (cache/) to improve response times.
• Configurable: Allows customization via configuration files (config.py, config/).

Installation

pip install llama-find
# Or install directly from GitHub for the latest version:
# pip install git+https://github.com/llamasearchai/llama-find.git

Usage

(Usage examples for the API client or web interface will be added here based on the final implementation.)

# Placeholder for Python client usage
# from llama_find import FinderClient

# client = FinderClient(config_path="path/to/config.yaml")
# results = client.search("your query here", engine="auto")
# print(results)

Architecture Overview

graph TD
    A[User / Client Application] --> B{API Layer (api.py / web_server.py)};
    B --> C{Search Agent (agents/)};
    C --> D[Core Processing (core.py)];
    D --> E{Search Engine Interface};
    E --> F[Search Engine 1];
    E --> G[Search Engine 2];
    E --> H[...];
    F --> I((External Search Service));
    G --> I;
    D --> J[Cache System (cache/)];
    D --> K[MLX Compatibility Layer (mlx_compat.py)];
    B --> L[Configuration (config.py, config/)];
    C --> L;
    D --> L;

    style D fill:#f9f,stroke:#333,stroke-width:2px
    style J fill:#ccf,stroke:#333,stroke-width:1px

1. Entry Point: Users or client applications interact via the API or web server.
2. API Layer: Handles incoming requests and routes them.
3. Search Agent: Manages the query lifecycle and interacts with core components.
4. Core Processing: Orchestrates the search, potentially leveraging MLX and caching.
5. Search Engine Interface: Abstracts interactions with different backend search engines.
6. Cache/Config: Caching improves speed; configuration allows customization.

Configuration

(Details on configuring search engines, API keys, caching, etc., will be added here.)

Development

Setup

# Clone the repository
git clone https://github.com/llamasearchai/llama-find.git
cd llama-find

# Install in editable mode with development dependencies
pip install -e ".[dev]"

Testing

pytest tests/

Contributing

Contributions are welcome! Please refer to CONTRIBUTING.md and submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.