​
What does Irona AI do?

Irona AI intelligently selects the best model for your query, significantly enhancing LLM output quality while minimizing costs and latency. Watch our demo video to see it in action.

​
Key Features

  • Maximize Performance: Irona AI surpasses Claude 3 Opus on major evaluation benchmarks. Our cold-start recommendations are grounded in extensive data from rigorous evaluation benchmarks and real-world scenarios.
  • Reduce Cost and Latency: Optimize quality, cost, and latency trade-offs to lower inference costs and achieve faster speeds. Irona AI selects the appropriate model more quickly than it takes an LLM to stream a single token.
  • Continuously Improve: Provide feedback on routing decisions to help Irona AI learn and refine a hyper-personalized routing algorithm tailored to your preferences and your application’s requirements.

​
Benchmark Performance

benchmark-performance By intelligent model selection, IronaAI consistently surpasses individual LLMs in accuracy and up to 10x cost savings.

​
Getting Started

Make your first IronaAI API request in minutes:
  1. Create an account at app.irona.ai
  2. Create a IronaAI API key
  3. Jump into the quickstart example

​
Installation

Requirements: Python 3.9+ 
pip install ironaai
We recommend creating and activating a virtual environment before installing the package to keep your dependencies isolated.

​
👍 TypeScript and REST Support

Not using Python? No problem! Use our TypeScript Client or directly call our REST API endpoint.