KILN-AI

What is KILN-AI?

KILN-AI is a free, intuitive desktop app for building AI systems and products, available on Windows, MacOS, and Linux. It also includes an MIT open‑source Python library and REST API for developers.

How to use KILN-AI?

Download the desktop app from kiln.tech/download, launch it, and follow the quickstart guide to connect any supported AI model (e.g., via Ollama, OpenAI, OpenRouter, AWS) and start building tasks.

Key features of KILN-AI

• Intuitive desktop apps for Windows, MacOS, and Linux
• Evaluations with state-of-the-art evaluators
• Zero-code fine-tuning for Llama, GPT-4o, and more
• Retrieval‑Augmented Generation (RAG) for knowledge integration
• Build agentic systems with multiple actors
• Connect powerful tools via Tools & MCP

Use cases of KILN-AI

• Evaluate and improve model quality on custom tasks
• Add domain knowledge to AI systems using document search (RAG)
• Fine‑tune models without writing code, then deploy them serverlessly
• Generate synthetic data for training or evaluation datasets
• Collaborate with teams using Git‑based version control for AI datasets

FAQ from KILN-AI

What models and providers does KILN-AI support?

KILN-AI supports over 100 tested models via Ollama, OpenAI, OpenRouter, Fireworks, Groq, AWS, any OpenAI‑compatible API, and more.

Does KILN-AI support MCP (Model Context Protocol) servers?

Yes, the Tools & MCP feature lets you connect powerful tools to your Kiln tasks.

Is KILN-AI free and open‑source?

The desktop apps are free, and the Python library and REST API are MIT open‑source.

How does KILN-AI handle data privacy?

KILN-AI runs locally on your computer and does not access your data. You bring your own API keys or use a local model via Ollama.

Can teams collaborate using KILN-AI?

Yes, KILN-AI provides Git‑based version control for AI datasets, enabling collaboration with QA, PM, and subject matter experts on data samples, evals, prompts, ratings, and issues.