Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1824 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Knowledge-Based Visual Question Answering | INFOSEEK | 18 | May 20, 2026 | ||
| Knowledge-Based Visual Question Answering | E-VQA | 18 | May 20, 2026 | ||
| Next-month anomaly forecasting | NCEP/NCAR Reanalysis 1 (held-out period 1980-2023) | 18 | May 20, 2026 | ||
| Earnings Forecasting | Swedish LISA register dataset Cohorts 1983–1985 (test) | 18 | May 20, 2026 | ||
| Image Segmentation | ADE | 18 | May 20, 2026 | ||
| Math Word Problem Solving | GSM8K (test) | 18 | May 20, 2026 | ||
| Multi-domain generalization | CORe50 Hard | 18 | May 20, 2026 | ||
| Multi-domain generalization | PACS | 18 | May 20, 2026 | ||
| Multi-domain generalization | Digits-DG | 18 | May 20, 2026 | ||
| Personal Finance | Personal Finance | 18 | May 20, 2026 | ||
| Targeted Attack | Synthetic Finance | 18 | May 20, 2026 | ||
| Targeted Attack | Enron Email | 18 | May 20, 2026 | ||
| Targeted Attack | HealthcareMagic-101 | 18 | May 20, 2026 | ||
| Untargeted data extraction attack | HealthcareMagic-101 | 18 | May 20, 2026 | ||
| Time Series OOD Generalization | UCIHAR, UniMiB-SHAR, EMG, Opportunity Aggregated | 18 | May 20, 2026 | ||
| Time Series OOD Generalization | Opportunity | 18 | May 20, 2026 | ||
| Time Series OOD Generalization | EMG | 18 | May 20, 2026 | ||
| Time Series OOD Generalization | UniMiB-SHAR | 18 | May 20, 2026 | ||
| Time Series OOD Generalization | UCIHAR | 18 | May 20, 2026 | ||
| Conversational Tool-use | τ²-Bench | 18 | May 28, 2026 | ||
| Binary Classification | 16 binary datasets | 18 | May 19, 2026 | ||
| Targeted evasion | WinAPI-AdvMal (test) | 18 | May 19, 2026 | ||
| Image Classification | ImageNet | 18 | May 19, 2026 | ||
| Perception-intensive Reasoning | MME-RealWorld-Lite (MRWL) | 18 | May 19, 2026 | ||
| Visual Search and Perception-intensive Reasoning | HR-Bench 8K | 18 | May 19, 2026 |