thunderbolt

Why we included this project

Thunderbolt is an open-source AI chat client built by the Thunderbird team, aimed at people who want to run model access on infrastructure they control. You bring the models: point it at a local Ollama or llama.cpp instance for free inference, or add API keys for any OpenAI-compatible provider in the settings. The client runs on phones, desktops, and in the browser, with Docker and Kubernetes deployment covered for on-prem setups. It's honest about being early, though: there's no public inference endpoint, and authentication and search currently depend on a backend you'd deploy yourself. If you're comfortable self-hosting and want one interface across several models with your data staying on your own machines, this is one to track as it matures.