Logical Intelligence announces Aleph AI agent benchmark scores

Logical Intelligence announced results for Aleph, its fully autonomous AI agent system for formal verification. The system recorded 99.4 percent on PutnamBench, 94 percent on VeriSoftBench, 100 percent on Verina and 32 out of 60 on LeanEval. A leaderboard graphic accompanied the announcement from the company.

Original post

PS#1426@PATRICKSHAFTO@LOGIC_INT

LO

Logical Intelligence@LOGIC_INT

Aleph, our fully autonomous AI agent system for formal verification, aced all major theorem proving benchmarks including PutnamBench, VeriSoftBench, and Verina

8:13 AM · May 14, 2026 View on X