New: RL environments scored against human competitors →

Benchmark · Optimisation

MathmoBench

Makes AI show its answer works, not guess it.

What it measures
Matching, scheduling, packing and allocation
Scored by
Executable proof or impossibility certificate
Headline figure
Scored in ms
Domain
Optimisation
Applies to
Optimisation software
Access
On request

How it is scored

Mathematical proof

A working solution, or a certificate that none exists. No proof, no score.

What it tests

Checkable answers, not guesses

  • Deliver a working solution or prove none exists. No proof, no score.
  • Scored in milliseconds, with unlimited new problems.
  • Matching, scheduling, packing and allocation: the core of real operations.

Experts behind it

Practitioners define what correct means for this benchmark. Paid, remote roles.

Evaluate your model on MathmoBench.