arcprize/arc-agi-3-benchmarking
42 22PythonMITUpdated 1 month ago
Our verdict: worth keeping an eye on
A ready‑made benchmarking harness for ARC‑AGI‑3; useful for evaluation but not required for day‑to‑day tooling
Our verdict: worth keeping an eye on
A ready‑made benchmarking harness for ARC‑AGI‑3; useful for evaluation but not required for day‑to‑day tooling