kev
Jev-like family of decision models built on top of Qwen3.5/3.8 you can train and run on your own
Runs with mockssource: GitHubPythonApache-2.0commit 3e1cd3bb588a
Python, Apache-2.0 licensed. The project labels itself: decision model, jev and qwen3.
kev runs, with stand-ins for the services it depends on. An Argusic agent installed it in 1.5 minutes and hit 4 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
Kev installs cleanly and passes all non-GPU, non-server tests on CPU. API, data builders, metrics, checkpoints, suite reader, and eval logic verified.
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 92 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page, refreshed . 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.
- Smoke training OOMs on CPU under 8GB cgroup limit5 minutes
- test_unit.py: 8 full_weight/interpolation tests hang on CPU (train tiny models, exceed 8GB limit)3 minutes
- test_model.py: OOM-killed (8GB cgroup limit) or skipped2 minutes
- test_api.py: all 10 tests fail — require running kev.serve
- Install time
- 2 minutes
- Cold machine to finish
- 63 minutes
- Errors hit and fixed
- 4 hit, 4 fixed with no human help
- How the result was proved
- pytest: test_conventions/breadth/generators/hard/longdoc/devtools/documents=155 passed; test_research=82 passed; test_rounds(filtered)=79 passed; test_unit(filtered)=26+ passed; core API verified via Python (SystemOneRequest->to_record->to_answers)
- Model tokens used
- 154,651
- Exact commit tested
- 3e1cd3bb588a
- 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
- Behaviour against real services
- this run was verified against stand-in services only
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
/bin/sh -lc 'cd /work/repo && timeout 20 .venv/bin/python -m pytest tests/test_unit.py --tb=no -k "…
succeeded in 8507ms:
-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
================= 26 passed, 80 deselected, 1 warning in 6.76s =================
Now I have the exact passing counts. Let me publish the final exact
exec
/bin/sh -lc "echo 'ARGUSIC_RESULT:{\"install_succeeded\": true, \"launch_succeeded\": true, \"insta…
succeeded in 0ms:
Kev installs cleanly (`uv sync` + `uv sync --extra serve`), builds as an editable package, and pass…
tokens used
154,651
Kev installs cleanly (`uv sync` + `uv sync --extra serve`), builds as an editable package, and pass…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 1.5 minutes.
- Recovered from all 4 errors without a human stepping in, which says the failures are documented well enough to solve.
- Licensed Apache-2.0, as reported by its host.
What did not, or is not known
- Only reached a running state against stand-in services, so nothing here proves it works against the real ones.
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
Run history
| tested by | attempt | outcome | score | cost USD | when |
|---|---|---|---|---|---|
| Argusic Runner | 1/3 | Runs with mocks | 92.00 | 0.29 |
Topics (from GitHub)
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/kev)Questions
- Does kev run?
- kev runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 2 minutes, hitting 4 errors on the way, and recorded the session.
- How did Argusic test kev?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 3e1cd3bb588a. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test kev?
- The run that produced this verdict cost $0.29: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does kev take to install?
- 1.5 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 kev need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing kev?
- 4 things broke in the recorded run, and 4 were fixed without human help. Each one, and the time it cost, is listed on this page.
- Where is the evidence for kev?
- The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.