Launching today
OnsetLab

OnsetLab

Local tool-calling AI agents with SLMs

19 followers

Build once, run anywhere. Your models, your tools, your machine. OnsetLab is an open-source framework for running tool-calling AI agents locally. Turn small language models into agents that can call real tools, work with your local environment, and stay under your control. No cloud lock-in, no hidden execution. Start in the playground, ship via Python, Docker, or vLLM.
OnsetLab gallery image
OnsetLab gallery image
OnsetLab gallery image
Free
Launch Team
Anima Playground
AI with an Eye for Design
Promoted

What do you think? …

Riyanshi Bohra
I built OnsetLab because I kept running into the same problem: most AI agent tools assume everything has to run in the cloud. For a lot of things I was building, that felt heavy, expensive, and harder to control than it needed to be. OnsetLab is my attempt to make AI agents work locally. You can take a small language model, let it call real tools, and run everything on your own machine. No hidden execution, no surprise data sharing. This is still early and very much a learning project, but I’m excited about where local and on-device AI is headed. I’d love to hear how others are thinking about local agents, tool-calling, or privacy-first AI. Thanks for checking it out!