Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2100 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| End-to-End Text Spotting | ICDAR 2003 (test) | 17 | Feb 26, 2026 | ||
| Human Pose Estimation | FLIC (test) | 17 | Feb 26, 2026 | ||
| Grasp Detection | Cornell Grasping Dataset (Image-wise split) | 17 | Feb 26, 2026 | ||
| Fine-grained Sentiment Classification | Stanford Sentiment Treebank (SST) (test) | 17 | Feb 26, 2026 | ||
| Classification | Peptides-func LRGB v1 (test) | 17 | Feb 26, 2026 | ||
| Molecular property prediction | ToxCast MoleculeNet (scaffold) | 17 | Feb 26, 2026 | ||
| Molecule Captioning | Mol-Instructions | 17 | Feb 26, 2026 | ||
| Clinical prediction | CRADLE | 17 | Feb 26, 2026 | ||
| Model-Human Comparison | model-vs-human (test) | 17 | Feb 26, 2026 | ||
| PAWP Prediction | Cardiac MRI (test) | 17 | Feb 26, 2026 | ||
| Binding affinity prediction | PDBBind core set 2016 (test) | 17 | Feb 26, 2026 | ||
| Classification | DVS128 Gesture | 17 | Apr 29, 2026 | ||
| Protein Design | CATH 4.2 (test) | 17 | Feb 26, 2026 | ||
| Classification | TCGA-MC3 (test) | 17 | Feb 26, 2026 | ||
| Classification | DisGeNet (test) | 17 | Feb 26, 2026 | ||
| Binary Classification | MoleculeNet SIDER DeepChem (test) | 17 | Feb 26, 2026 | ||
| Binary Classification | MoleculeNet BACE DeepChem (test) | 17 | Feb 26, 2026 | ||
| Binary Classification | MoleculeNet Tox21 DeepChem (test) | 17 | Feb 26, 2026 | ||
| Binary Classification | MoleculeNet BBBP DeepChem (test) | 17 | Feb 26, 2026 | ||
| Retrosynthesis | USPTO-50K unknown reaction types (test) | 17 | Feb 26, 2026 | ||
| Retrosynthesis | USPTO-50K Reaction types given as prior (test) | 17 | Feb 26, 2026 | ||
| Question Answering | ARC-Challenge | 17 | Feb 26, 2026 | ||
| Job Shop Scheduling | Lawrence's benchmark Avg 40 instances | 17 | Feb 26, 2026 | ||
| Job Shop Scheduling | Demirkol's benchmark Avg | 17 | Feb 26, 2026 | ||
| Point cloud registration | ModelLoNet 40 (test) | 17 | Feb 26, 2026 |