kapitsa social preview
inference-server713Apache 2.0

kapitsa

Blazing fast Apple Silicon inference engine optimized for Qwen & Nemotron. OpenAI + Anthropic API compatible. Built by folks from nace.ai

Updated Nov 9, 2026
Website GitHub
Platforms
macos, linux, windows
Pricing
free-open-source
Status
active
License
Apache 2.0

What it does

Core capabilities at a glance

  • Apple Silicon
  • Clover
  • Coding Agents
  • Decision Models
  • Drex
  • Drex MAX
  • Fast Local Inference
  • Flint

Deep dive

The full breakdown - performance, comparisons, and setup

kapitsaLink to this section

kapitsa is a local inference server - Blazing fast Apple Silicon inference engine optimized for Qwen & Nemotron. OpenAI + Anthropic API compatible. Built by folks from nace.ai.

OverviewLink to this section

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, Pi, Hermes, aider or Goose 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.

Limits: Apple silicon only: no Linux or Windows. Flint, the four text-only Nemotrons and the 35B's '16gb' build read text only, not images (which models read images).

Kapitsa has the fastest decode in 33 of the 38 model and Mac pairs measured against nine engines (how each number was measured).

Each cell reads prefill / decode in tokens per second, from a 4K-token prompt and a 512-token answer at T 0.6 (Flint at its card's T 1.0). The M5 Pro's Flash-Next 180B Q2 row is a 256-token answer at T 0 on that run's own 4,074-token prompt, and its Kapitsa column is Kapitsa with KFlash against plain Splash 1.5.0: a tie, the two engines' ranges overlap (record). Footnotes, run dates and flags: docs/benchmarks.md.

kapitsa is open-source, written primarily in C++, with 713 GitHub stars under the Apache 2.0 license. The latest release is v0.2.32 (2026-11-08).

Key capabilitiesLink to this section

From the project's documentation:

  • A 16K follow-up after a restart starts in 0.16 s instead of 17 s. Conversations stay on the SSD (--disk-cache).

InstallLink to this section

A quick way to get started (always check the official docs for the latest):

brew install nace-ai/tap/kapitsa

How it fits a local-AI stackLink to this section

kapitsa runs on your own hardware, so pair it with a model and a GPU sized to your needs. Use the VRAM calculator to pick a model that fits your card, and see what you can run for hardware guidance. Related local inference servers in the directory:

SourcesLink to this section

Stats from GitHub, 2026-11-09.

Frequently asked

Quick answers to common questions

What is kapitsa?

kapitsa is a inference-server tool for local AI workloads. Blazing fast Apple Silicon inference engine optimized for Qwen & Nemotron. OpenAI + Anthropic API compatible. Built by folks from nace.ai

Is kapitsa free and open source?

Yes, kapitsa has 713 GitHub stars and is licensed under Apache 2.0. You can self-host it for free on macos, linux, windows.

What platforms does kapitsa support?

kapitsa runs on macos, linux, windows.

What hardware do I need for kapitsa?

The hardware requirements depend on which models you run. Check our hardware directory for compatible GPUs and systems. kapitsa has 713 GitHub stars and an active community.

Does kapitsa support GPU acceleration?

kapitsa supports GPU acceleration via CUDA, Metal, or Vulkan depending on your platform. For the best performance, pair it with an NVIDIA RTX 4090 or 5090.

What are the best alternatives to kapitsa?

Popular alternatives include other inference-server tools in our directory. Browse our full collection at /tool for comparisons, community reviews, and benchmark data to find the right fit for your workflow.

How much does kapitsa cost?

kapitsa is free-open-source. It is completely free and open source to self-host.

Pairs well with

Complementary tools, models, and hardware