Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1782 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Protein fitness prediction | ProteinGym | 19 | Feb 26, 2026 | ||
| Dual-target drug design | 12,917 pairs of targets dual-target setting (test) | 19 | May 26, 2026 | ||
| Tile-level classification | PCam (test) | 19 | Feb 26, 2026 | ||
| Continuous Control | D4RL Walker2d medium | 19 | Jun 9, 2026 | ||
| Global Optimization | Type U various geometries synthetic | 19 | Feb 26, 2026 | ||
| Per time-step regression | Walker2D | 19 | Feb 26, 2026 | ||
| Global Optimization | F6 benchmark function | 19 | Jun 4, 2026 | ||
| Graph Classification | RDT-B (test) | 19 | Feb 26, 2026 | ||
| Word-level prediction | PTB word-level (test) | 19 | Feb 26, 2026 | ||
| Time Series Classification | CMUSUBJECT16 (test) | 19 | Feb 26, 2026 | ||
| Image Classification | MNIST-10 (test) | 19 | Feb 26, 2026 | ||
| Material generation | MP-20 (test) | 19 | May 12, 2026 | ||
| Force prediction | MD17 revised (test) | 19 | Feb 26, 2026 | ||
| Jet Tagging | Quark-Gluon Tagging (test) | 19 | May 19, 2026 | ||
| Regression | FreeSolv (scaffold) | 19 | Feb 26, 2026 | ||
| Molecular energy prediction | QM9 | 19 | Feb 26, 2026 | ||
| Graph Retrieval | FM (test) | 19 | Feb 26, 2026 | ||
| User-Generated Content Video Quality Assessment | YouTube-UGC original (test) | 19 | Feb 26, 2026 | ||
| Pansharpening | WorldView-2 reduced-resolution 256 × 256 subscene | 19 | Feb 26, 2026 | ||
| Real-world Understanding | R-Bench | 19 | Feb 27, 2026 | ||
| 3D Object Classification | ScanObjectNN OBJ_ONLY v1 | 19 | Feb 26, 2026 | ||
| Reflective Object Reconstruction | Glossy Synthetic | 19 | Mar 31, 2026 | ||
| CUA classification | 3D uterine ultrasound dataset | 19 | Feb 26, 2026 | ||
| Long-form Question Answering | ASQA | 19 | Apr 10, 2026 | ||
| Semantic Segmentation | PST900 (evaluation) | 19 | Feb 26, 2026 |