Argusic could not verify that kapitsa runs. Argusic installed and launched it on a clean machine in 0 minutes, hitting 1 error on the way, and recorded the session.

How did Argusic test kapitsa?

On a fresh, disposable machine, with every command recorded and the repository pinned at commit e21f08b45fde. 2 attempts are recorded, and the full method is on the methodology page.

What did it cost to test kapitsa?

The run that produced this verdict cost $0.01: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.

How long does kapitsa take to install?

0 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.

Does kapitsa need a GPU?

Not to start. Argusic reached a running state on a machine with no GPU.

What goes wrong when installing kapitsa?

1 thing broke in the recorded run, and 1 were fixed without human help. Each one, and the time it cost, is listed on this page.

Where is the evidence for kapitsa?

All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.

kapitsa

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

Could not verifysource: GitHubC++Apache-2.0commit e21f08b45fde

C++, Apache-2.0 licensed. The project labels itself: apple silicon, clover, coding agents, decision models, drex, drex max, fast local inference and flint.

Argusic could not get kapitsa running. An Argusic agent installed it in 0 minutes and hit 1 error and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

Nothing builds or runs. The project is Apple Silicon (arm64, macOS 14+) only and is unreachable from this x86_64 Linux container.
What the agent ended up with on a clean machine, in its own words. How this is measured

Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 10 of 100 (the mean of this project's run scores).

At a glance

verdict: Could not verify
Argusic Score: 10 / 100
cost of the verifying run: $0.01 (measured)
recorded runs: 2
last tested: 2026-11-09
stars: 713
forks: 0
open issues: 0
watchers: 0
created: -
last push: -

Subject data from GitHub, linked at the top of this page. Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.

What was measured

From the run that decided this verdict. Every line is a recorded value, not an estimate.

What went wrong, in the agent's own words

The agent's own words, worst first, never rewritten into advice.

Kapitsa requires Apple Silicon (arm64) and macOS 14+ with Xcode Command Line Tools. The environment is x86_64 Linux, which has no Objective-C++ compiler (cc1objplus), no Metal GPU framework, and no Objective-C runtime. CMake configure fails at the OBJCXX compiler detection step.

Install time: under a minute
Cold machine to finish: under a minute
Errors hit and fixed: 1 hit, 1 fixed with no human help
How the result was proved: cmake -S . -B build -DCMAKE_BUILD_TYPE=Release failed with 'The Objective-C++ compiler is not able to compile a simple test program' at the project() call. uname returned Linux x86_64. install.sh checks for Darwin + arm64 on line 1. Verified the 4 Python-only tests passed (policy/docs checks only).
Model tokens used: 37,721
Exact commit tested: e21f08b45fde
Container image: sha256:cdd920bce783

What was not measured

The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.

Peak GPU memory: this ran on a CPU machine, so there was no GPU to sample
Behaviour over a long run: each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
Whether it works at all: this run verified nothing, so this page makes no claim either way

The end of the recording

The last lines of the terminal session that decided this verdict, exactly as they were recorded.

exec
/bin/sh -lc "# Check if we can try anything else - does the README mention any cross-platform or Do…
cd /work/repo && rg -l \"docker\\|container\\|linux\\|x86\\|cross\" --type md 2>/dev/null | head -1…
 succeeded in 0ms:
exec
/bin/sh -lc "# Check if there's a Docker environment for Kapitsa
rg -l \"Docker\\|docker\\|container\" /work/repo/ --include '*.md' --include '*.yml' --include '*.y…
 succeeded in 0ms:
No Docker setups, no cross-platform support. This is purely Apple Silicon native.
tokens used
37,721
No Docker setups, no cross-platform support. This is purely Apple Silicon native.

Replay the whole session, every command from a clean machine to this point.

How it was tested

One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.

Strengths and limits

Measured facts, not opinions. How this is written.

What went well

Installed in 0 minutes, faster than the median of the 6 comparable projects Argusic has measured.
Recovered from all 1 error without a human stepping in, which says the failures are documented well enough to solve.
Licensed Apache-2.0, as reported by its host.
Measured 2 times, so the result is not a one-off.

What did not, or is not known

An agent with no time limit did not get it running.

Of the 7 apple-silicon projects Argusic has installed and timed, kapitsa was the 1st fastest to reach a running state, and 2 of 7 reached one at all.

Also tested, in the same area

Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.

Run history

tested by	attempt	outcome	score	cost USD	when
Argusic Runner	2/3	Did not run	0.00	0.03	2026-11-09 07:43 UTC
Argusic Runner	1/3	Did not run	20.00	0.01	2026-11-09 07:17 UTC

Topics (from GitHub)

apple-silicon clover coding-agents decision-models drex drex-max fast-local-inference flint llama-cpp llm-inference local-llm macos metal mlx nemotron ollama openai-api qwen slm speculative-decoding

Embed the badge

Markdown for the project README. It links back here; terms on the terms page.

[![Tested by Argusic](https://argusic.com/badge/kapitsa.svg)](https://argusic.com/subject/kapitsa)

Questions

Does kapitsa run?: Argusic could not verify that kapitsa runs. Argusic installed and launched it on a clean machine in 0 minutes, hitting 1 error on the way, and recorded the session.
How did Argusic test kapitsa?: On a fresh, disposable machine, with every command recorded and the repository pinned at commit e21f08b45fde. 2 attempts are recorded, and the full method is on the methodology page.
What did it cost to test kapitsa?: The run that produced this verdict cost $0.01: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does kapitsa take to install?: 0 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
Does kapitsa need a GPU?: Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing kapitsa?: 1 thing broke in the recorded run, and 1 were fixed without human help. Each one, and the time it cost, is listed on this page.
How does kapitsa compare with the alternatives?: Of the 7 apple-silicon projects Argusic has installed and timed, kapitsa was the 1st fastest to reach a running state, and 2 of 7 reached one at all.
Where is the evidence for kapitsa?: All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.