kapitsa
Blazing fast Apple Silicon inference engine optimized for Qwen & Nemotron. OpenAI + Anthropic API compatible. Built by folks from nace.ai
kapitsa launched on 16 October 2026. It ranked #119 on Daily Launch on 8 November 2026 with a Launch Score of 10, from +31 stars on 8 Nov on GitHub. It moved down 3 places from the day before.
What is kapitsa?
Kapitsa is an inference engine for Apple silicon that runs Qwen, Nemotron and Recursive's Flint models locally, built for coding agents. It starts opencode, Claude Code, Codex or pi on a local model, and serves OpenAI and Anthropic APIs. Requirements: a Mac with Apple silicon (M1 or newer) and macOS 14 Sonoma or later.
From its README on GitHub
kapitsa is a repo by nace-ai on GitHub. It is written in C++ and released under the Apache-2.0 license.
How to get kapitsa
brew install nace-ai/tap/kapitsaFrom its README on GitHub
More steps and the source are in its repository on GitHub.