jev

jevos: an open-source alternative to Jev and Microsoft-Decision-1. Yes/no decisions, choices and scores with probabilities, on your laptop's CPU.

Runssource: GitHubhomepageC++MITcommit eb74cf78e537

C++, MIT licensed. The project labels itself: ai, classification, cpu inference, decision model, edge ai, gguf, jev and jev alternative.

jev runs. An Argusic agent installed it in 19 minutes and hit 2 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

The jev project installs with uv, the llama.cpp runtime and a pinned GGUF model download with sha256 verification, and the HTTP server (port 8017) serves real noul decisions over /v1/systemone plus /health, /v1/models and /dino, all confirmed with curl status codes and JSON responses in this 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 60 of 100 (the mean of this project's run scores).

At a glance

verdict
Runs
Argusic Score
60 / 100
cost of the verifying run
$0.05 (measured)
recorded runs
2
last tested
stars
970
forks
150
open issues
5
watchers
14
size
18 MB
created
last push

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.

Time lost to each failure while testing jev, drawn from the recorded run
Every bar is a failure the agent recorded, with the time it cost. Drawn from this run, not from anywhere else.

What went wrong, in the agent's own words

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

  • README quickstart points to a jevos release asset (jevos-q4_k_m.gguf) that returns 404 on GitHub4 minutes
  • uv was not installed in the container2 minutes
Install time
19 minutes
Cold machine to finish
7 minutes
Errors hit and fixed
2 hit, 2 fixed with no human help
How the result was proved
uv sync exited 0 (16 packages resolved); 'uv run jev serve --model spark-1.7b --quant q4_k_m --device cpu --threads 4' started and GET /health returned HTTP 200 with status ready; POST /v1/systemone with the README example returned HTTP 200 with noul 0.9496128141559874; 'jev decide' on a native request exited 0 and wrote full decision JSON; /v1/models and /dino both returned HTTP 200
Model tokens used
191,451
Exact commit tested
eb74cf78e537
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

Watch the session

The whole run, from a clean machine, at the speed it actually happened. Nothing is cut and nothing is sped up.

A recording of attempt 1, run , the attempt that decided the verdict above. The terminal recording it was made from is the evidence, stored with a sha256 fingerprint; the video is a rendering of it. Nothing loads from the video host until you press play.

The end of the recording

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

5. **Real inference** — `POST /v1/systemone` with the README's example returned HTTP 200 with `{"mo…
6. **CLI** — `jev decide --model spark-1.7b --quant q4_k_m --device cpu --threads 4` on a native-fo…
7. **Assets** — `ffmpeg -v error` decoded `assets/dino_run.gif` and both PNGs cleanly.
8. **One doc mismatch** (not a failure): README says `choice`/`score` are refused with 422, but the…
Total elapsed: ~19 minutes of work (install ~4 min, downloads ~3 min, runs ~10 min).
exec
/bin/sh -lc 'cd /work/repo && export PATH="$HOME/.local/bin:$PATH" && uv run jev devices 2>&1 | pyt…
 succeeded in 0ms:
installed_runtimes: ['vulkan']
devices: ['CPU']
tokens used
191,451

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

  • Reached a running state on a clean machine, with the session recorded.
  • Installed in 19 minutes.
  • Recovered from all 2 errors without a human stepping in, which says the failures are documented well enough to solve.
  • Ran without a GPU, so it does not need one to start.
  • Licensed MIT, as reported by its host.
  • Measured 2 times, so the result is not a one-off.

What it is a reasonable choice for

  • Trying it on a laptop or a small server: it reached a running state without a GPU.

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 byattemptoutcomescorecost USDwhen
Argusic Runner2/3Did not run20.000.00
Argusic Runner1/3Runs100.000.05

Topics (from GitHub)

aiclassificationcpu-inferencedecision-modeledge-aiggufjevjev-alternativellama-cppllmllm-inferencelocal-ailocal-llmmicrosoft-decisionmicrosoft-decision-1offlineon-device-aiopen-source-alternativeself-hostedyes-no

Embed the badge

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

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

Questions

Does jev run?
Yes. jev runs. Argusic installed and launched it on a clean machine in 19 minutes, hitting 2 errors on the way, and recorded the session.
How did Argusic test jev?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit eb74cf78e537. 2 attempts are recorded, and the full method is on the methodology page.
What did it cost to test jev?
The run that produced this verdict cost $0.05: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does jev take to install?
19 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 jev need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing jev?
2 things broke in the recorded run, and 2 were fixed without human help. Each one, and the time it cost, is listed on this page.
Where is the evidence for jev?
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.

Discussion