minimind
🧠Train a 64M-parameter LLM from scratch in just 2h!
Runssource: GitHubPythonApache-2.0commit f659b55761b7
Python, Apache-2.0 licensed. The project labels itself: artificial intelligence and large language model.
minimind runs. An Argusic agent installed it in 5 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 MiniMind-3 63.91M model runs end-to-end on CPU: CLI inference produces coherent responses, the OpenAI-compatible FastAPI server serves chat completions (streaming and non-streaming) with HTTP 200, and the Streamlit web interface loads successfully on port 8501.
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 100 of 100 (the mean of this project's run scores).
At a glance
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.
- ujson==5.1.0 failed to build from source: missing Python.h (no python3-dev headers); no root to install system packages2 minutes
- System Python is externally managed (PEP 668); pip install to system Python fails1 minute
- Install time
- 5 minutes
- Cold machine to finish
- 12 minutes
- Errors hit and fixed
- 2 hit, 2 fixed with no human help
- How the result was proved
- eval_llm.py with --load_from ./minimind-3/gongjy/minimind-3 --device cpu generated valid responses on all 8 prompts; curl -X POST http://localhost:8998/v1/chat/completions returned 200 with valid completion JSON; curl http://localhost:8501 returned HTTP 200
- Model tokens used
- 161,034
- Exact commit tested
- f659b55761b7
- 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
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
lrwxrwxrwx 1 runner runner 7 Oct 5 07:16 /work/repo/env/bin/python -> python3 2026-10-05 07:18:39.343219926 +0000 exec /bin/sh -lc 'date +%s; date; stat -c %y /work/repo/env/ 2>/dev/null' in /work/repo succeeded in 0ms: 1791185276 Mon Oct 5 07:27:56 UTC 2026 2026-10-05 07:18:39.343219926 +0000 Total session time is about 10 minutes. Let me finalize the results. tokens used 161,034 Total session time is about 10 minutes. Let me finalize the results.
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 5 minutes, faster than the median of the 4 comparable projects Argusic has measured.
- 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 Apache-2.0, as reported by its host.
What did not, or is not known
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
Of the 5 artificial-intelligence projects Argusic has installed and timed, minimind was the 2nd fastest to reach a running state, and 4 of 5 reached one at all.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
- A quick evaluation: it was running 5 minutes after the clone.
Run history
Topics (from GitHub)
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/minimind)Questions
- Does minimind run?
- Yes. minimind runs. Argusic installed and launched it on a clean machine in 5 minutes, hitting 2 errors on the way, and recorded the session.
- How did Argusic test minimind?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit f659b55761b7. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test minimind?
- The run that produced this verdict cost $0.09: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does minimind take to install?
- 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 minimind need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing minimind?
- 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.
- How does minimind compare with the alternatives?
- Of the 5 artificial-intelligence projects Argusic has installed and timed, minimind was the 2nd fastest to reach a running state, and 4 of 5 reached one at all.
- Where is the evidence for minimind?
- 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.