Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1467 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Privacy-preserving neural network inference | AlexNet (inference) | 21 | Mar 16, 2026 | ||
| De Novo Design | 10 de novo shared-core domains | 21 | Mar 16, 2026 | ||
| Adversarial Attack | ImageNet 1k (test) | 21 | Mar 16, 2026 | ||
| Deraining | AIO-IR Deraining | 21 | Jun 23, 2026 | ||
| Aggregate Data Analysis Performance | DABench, TableBench, and BIRD Combined | 21 | Mar 16, 2026 | ||
| Table Reasoning | TableBench | 21 | Mar 16, 2026 | ||
| Data Analysis | DABench | 21 | Mar 16, 2026 | ||
| Person Re-identification | Average (Market1501, CUHK03, MSMT17) Protocol-1 (test) | 21 | Mar 16, 2026 | ||
| Person Re-identification | MSMT17 C2+C3+M→MS Protocol-1 (test) | 21 | Mar 16, 2026 | ||
| Person Re-identification | Market1501 MS+C2+C3→M Protocol-1 (test) | 21 | Mar 16, 2026 | ||
| Federated Learning | CIFAR-100 | 21 | Mar 13, 2026 | ||
| Federated Learning | CIFAR-10 | 21 | Mar 13, 2026 | ||
| 4D scene reconstruction | iPhone | 21 | Mar 13, 2026 | ||
| Recommendation | Amazon Video | 21 | Jun 16, 2026 | ||
| Seizure Frequency Extraction | Real 300 1.0 (Purist) | 21 | Mar 13, 2026 | ||
| Speaker Verification | VoxCeleb1 hard (H) | 21 | Jun 23, 2026 | ||
| Speaker Verification | VoxCeleb1 extended | 21 | Jun 23, 2026 | ||
| Multi-view 3D reconstruction | MV-dVRK | 21 | Mar 13, 2026 | ||
| Multi-view 3D reconstruction | 4D-DRESS | 21 | Mar 13, 2026 | ||
| Synthetic Graph Generation | Tree Dataset | 21 | May 18, 2026 | ||
| Instruction-tuning | UNI | 21 | Mar 12, 2026 | ||
| Instruction-tuning | SelfInst | 21 | Mar 12, 2026 | ||
| Instruction-tuning | SNI | 21 | Mar 12, 2026 | ||
| Instruction-tuning | Dolly | 21 | Mar 12, 2026 | ||
| Video Reasoning Quality Assessment | UrbanVideo and VSI-Bench Clean | 21 | Mar 12, 2026 |