Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1098 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Object-level hallucination evaluation | Obj-Hal | 25 | Jul 1, 2026 | ||
| AI-generated video detection | GenVidBench matched 27k | 25 | Jul 1, 2026 | ||
| Deepfake Detection | DM-OS (Diffusion Model-Open Source) Out-of-domain (test) | 25 | Jun 30, 2026 | ||
| Deepfake Detection | DM-CS Diffusion Model-Closed Source Out-of-domain (test) | 25 | Jun 30, 2026 | ||
| Anomaly Detection | aloi | 25 | Jun 30, 2026 | ||
| Anomaly Detection | Average | 25 | Jun 30, 2026 | ||
| Anomaly Detection | aloi | 25 | Jun 30, 2026 | ||
| Nudity Concept Removal | UnlearnAtk | 25 | Jun 30, 2026 | ||
| Commonsense Reasoning | HellaSwag | 25 | Jul 7, 2026 | ||
| Multi-robot exploration | KTH 8 campus floors, 8 seeds | 25 | Jun 30, 2026 | ||
| Multi-robot exploration | HouseExpo 16 homes, 20 seeds | 25 | Jun 30, 2026 | ||
| Robot Video Diagnosis | RoboGazeBench (test) | 25 | Jun 30, 2026 | ||
| Feature Attribution | Synthetic VAR Data | 25 | Jun 29, 2026 | ||
| Anomaly recognition | Distractor dataset (d) | 25 | Jun 26, 2026 | ||
| Anomaly recognition | Plain dataset (p) | 25 | Jun 26, 2026 | ||
| Medical Reasoning | Medical Reasoning Evaluation Suite m23k | 25 | Jun 26, 2026 | ||
| Fault Detection | CCGT (test) | 25 | Jun 26, 2026 | ||
| Fault detection | Fault detection dataset (test episodes) | 25 | Jun 26, 2026 | ||
| EEG Classification | EEG state classification dataset 4 states | 25 | Jun 26, 2026 | ||
| Object Detection | HRPlane v2 (test) | 25 | Jun 25, 2026 | ||
| Video Face Restoration | VFHQ (test) | 25 | Jul 8, 2026 | ||
| OCR Benchmarking | OCRBench | 25 | Jun 24, 2026 | ||
| Fake and manipulated image detection | Synthbuster | 25 | Jul 8, 2026 | ||
| MCQ-Neg | NegBench | 25 | Jun 24, 2026 | ||
| Negative Retrieval | NegBench | 25 | Jun 24, 2026 |