Measured benchmarks
MMLU-Pro (TIGER-Lab)
Published by TIGER-Lab (University of Waterloo). Source page ↗ · Data we read (file download) ↗
Submissions mix TIGER-Lab-measured and self-reported rows; the display name says which.
Facts
| Fact | Value | Where it comes from |
|---|---|---|
| Who runs the evaluation | a university group | curated by aiqbench, reviewed 2026-10-09 |
| Test set exposure | public | curated by aiqbench, reviewed 2026-10-09 |
| Data licence | Apache-2.0 | the board's attribution record |
| Newest published result | not published | the ingested rows |
| Last pulled by aiqbench | 9 hours ago · every 12 h | the sync log |
| Models on this axis | 260 | the ingested rows |
| Rows with a published interval | none published | the ingested rows |
| Headroom at the top | Best 91.2%, 8.8% below the ceiling; top 10 within 3.8%. | computed from the ingested rows |
| Axes it tracks closely |
| Spearman ρ over shared models (see Clusters) |
Where it appears
Not in any built-in template; selectable in the console.
View this axis alone in the console →Attribution
Source: MMLU-Pro Leaderboard (TIGER-Lab, University of Waterloo) — https://huggingface.co/datasets/TIGER-Lab/mmlu_pro_leaderboard_submission. Licensed under Apache 2.0. Modified by aiqbench: we ingest the 'results.csv' file from the HF dataset root, rescale each row's Overall accuracy from 0-1 to percent, and preserve every disciplinary sub-score (Biology, Business, Chemistry, Computer Science, Economics, Engineering, Health, History, Law, Math, Philosophy, Physics, Psychology, Other) plus the Model Size and Data Source (TIGER-Lab vs Self-Reported) verbatim in each entry's metadata. Rows self-reported by model authors are labelled '(self-reported)' in the display name so readers can see the provenance without opening the details panel.