Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2628 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Sequential Recommendation | Yelp | 14 | May 28, 2026 | ||
| Sequential Recommendation | Beauty | 14 | May 28, 2026 | ||
| Natural Language Inference | GLUE and SNLI v1 (test val) | 14 | May 28, 2026 | ||
| Knowledge Editing | MQuAKE-T 3-hop | 14 | May 28, 2026 | ||
| Knowledge Editing | MQuAKE-T 2-hop | 14 | May 28, 2026 | ||
| General Multimodal Evaluation | MMMU (val) | 14 | May 28, 2026 | ||
| Text-to-Video | MS-COCO random 1,000 prompts 2017 (test) | 14 | May 28, 2026 | ||
| Math Reasoning | MATH 500 | 14 | May 28, 2026 | ||
| Mathematical Reasoning | AMC 24 | 14 | May 28, 2026 | ||
| Sim-to-Real Acoustic Parameter Prediction | HAA Dampened 8:2 ratio (test) | 14 | May 28, 2026 | ||
| Sim-to-Real Acoustic Parameter Prediction | HAA Hallway 8:2 ratio (test) | 14 | May 28, 2026 | ||
| Visual Question Answering | GQA | 14 | Jul 1, 2026 | ||
| Safety Evaluation | Q-LatHarmful | 14 | May 28, 2026 | ||
| Safety Evaluation | LatHarmful | 14 | May 28, 2026 | ||
| Safety Evaluation | I-BeaverTails | 14 | May 28, 2026 | ||
| Code Generation | HumanEval with PEP 8 compliance n=164 (test) | 14 | May 28, 2026 | ||
| Wikipedia NPOV rewriting | Wikipedia NPOV rewriting (n=107) | 14 | May 28, 2026 | ||
| Rule-governed Reasoning | RuleArena NBA | 14 | Jun 4, 2026 | ||
| Clinical Diagnosis | RJUA (OOD) | 14 | May 28, 2026 | ||
| Clinical Diagnosis | MedXpertQA (OOD) | 14 | May 28, 2026 | ||
| Clinical Diagnosis | MedQA (OOD) | 14 | May 28, 2026 | ||
| Clinical Diagnosis | MedDDx-Plus (In-domain) | 14 | May 28, 2026 | ||
| Multi-step mathematical reasoning | GSM8K (test) | 14 | May 28, 2026 | ||
| Multi-step mathematical reasoning | MMLU-Pro (held-out test) | 14 | May 28, 2026 | ||
| Multi-step mathematical reasoning | MMLU (test) | 14 | May 28, 2026 |