Benchmark Almanac

67AI Lab

We track 2 benchmarks that 67AI Lab made or helped make.

0 in use 0 with public code 2 with a leaderboard here

Benchmarks

Name Kind Released Best score Status
67AI Lab L2 nuclear medicine report structuring Models extract findings from 300 synthetic PET/CT and SPECT reports into a fixed form. They also flag conflicting values in 17 reports. Medicine Nov 2026 1.00 Not in cards
67AI Lab L3 nuclear medicine guideline questions Models answer questions using eight retrieved passages from nuclear medicine guidelines. They must abstain when the passages lack an answer. Medicine Nov 2026 90.8% Not in cards