connecting to the lab…

A home cluster that tries to prove theorems.

Small AI models running on secondhand graphics cards write proofs. The Lean 4 proof checker reads every one and says yes or no. It never takes the model's word for anything.

Right now

The lab, live

These numbers come from the cluster itself and refresh every 15 seconds. The lab pushes them out; nothing on the internet can reach in.

Proofs accepted
–
 
Attempts checked
–
 
GPUs working
–
 
Hottest GPU
–
slows itself at 96°

The graphics cards

Each card: temperature now, load bar, and temperature over the last 3 hours on a fixed 20 to 100 °C scale. Hover a line for values.

Power drawn by the GPUs, last 3 hours
–
Power station
–
battery
–
solar coming in
–
going out

The cluster runs behind a battery power station fed by solar panels, which is why leaving a proof search running for weeks is affordable.

The idea

Guess, check, repeat. For as long as it takes.

A language model can write something that looks like a proof and is wrong. Lean cannot be talked into anything: a proof either compiles or it does not. Put the two together and a wrong answer costs nothing but time.

So the lab takes one small lemma at a time, asks a local model for a proof, and hands it to Lean. A failure comes back with the exact error, and the model tries again with that error in hand.

Below is the checker answering four real submissions from the lab's test suite: a good proof, a false statement, a placeholder, and an attempt to cheat with a made-up axiom.

lean-worker

      
The loop

Five roles, one rule

Nothing counts as proved unless Lean compiled it with no placeholders and no added axioms.

Architectsplits goals into lemmas Foremanqueue, retries, alerts Local modelspropose a proof Lean 4compiles it: yes or no Ledgerverified proofs in Git pass fail: the error text goes back for another try out of retries: split the lemma smaller
Progress

Where the build stands

  1. 1

    Lean worker

    A container that compiles any Lean file against Mathlib and answers in about 3 seconds.

    ✔ done
  2. 2

    Model workers

    Model servers are up on every GPU host. Next: benchmark candidate models on 20 easy lemmas and keep the winner.

    ◐ in progress
  3. 3

    Orchestration

    The unattended loop: queue, retry limits, timeouts, a daily summary. Goal: 48 hours without stalling.

    ○ next
  4. 4

    Real targets

    A backlog of results that are known but not yet in Mathlib, Lean's public maths library.

    ○ planned
  5. 5

    Open to contributors

    Public lemma backlog and proofs, with every outside submission verified away from the home network.

    ○ planned
Honest scope

What this will and will not do

It will not solve a Millennium Prize problem

Time helps with search. It does not make a small model smarter. Hard steps have to be broken down by a stronger model or a person first.

It can grind small lemmas

Many small, well-defined statements, tried over and over, with a referee that rejects every wrong answer for free.

The statement still needs a human

A proof that compiles shows the statement was proved, not that the statement says what you meant. Every lemma is reviewed before it enters the queue.