Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2199 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Organ-presence classification | BTCV | 16 | Apr 3, 2026 | ||
| Chart2Code | ChartX | 16 | Apr 3, 2026 | ||
| Long-term Memory Question Answering | LongMemEval-S (500 questions) | 16 | Jun 16, 2026 | ||
| Code Generation | LeetCode without write access | 16 | Apr 3, 2026 | ||
| Reward Hacking | LeetCode with write access (test) | 16 | Apr 3, 2026 | ||
| Question Answering | NQOpen | 16 | Apr 3, 2026 | ||
| Calibration | PopQA | 16 | Apr 3, 2026 | ||
| Annotation-free closed-set semantic segmentation | nuScenes (val) | 16 | Apr 3, 2026 | ||
| Proficiency Estimation | Ego-Exo4D | 16 | Jun 2, 2026 | ||
| Egocentric Text Retrieval | Ego-Exo4D | 16 | Apr 30, 2026 | ||
| Classification | 18 UEA & UCR datasets Average (test) | 16 | Apr 3, 2026 | ||
| Classification | UEA & UCR 18 datasets 50% Missing (test) | 16 | Apr 3, 2026 | ||
| Classification | 18 UEA & UCR datasets Regular (test) | 16 | Apr 3, 2026 | ||
| Crisis response generation | 16 | Apr 3, 2026 | |||
| Classification | CrossFit cls-23 | 16 | Apr 3, 2026 | ||
| Question Answering | CommonsenseQA ConceptNet (20% test) | 16 | Apr 3, 2026 | ||
| Zero-shot Text Evaluation | Text Benchmarks ARC-e, ARC-c, BoolQ, HS, OBQA, PIQA, WG, MMLU | 16 | Apr 2, 2026 | ||
| Prompt injection defense | WASP | 16 | Apr 10, 2026 | ||
| Competitive Programming | LiveCodeBench v6 | 16 | Apr 2, 2026 | ||
| Video Understanding | Video-MME | 16 | Jul 7, 2026 | ||
| Object Probing | POPE | 16 | Apr 2, 2026 | ||
| Multimodal Understanding | MME | 16 | Apr 2, 2026 | ||
| Referring Expression Segmentation | RefCOCO, RefCOCO+, and G-Ref (average) | 16 | Apr 7, 2026 | ||
| LLM Inference Scheduling | LMSYS-Chat-1M | 16 | Jun 29, 2026 | ||
| LD-weighted PDF reconstruction | three-mode NLS Outside only distribution (test) | 16 | Apr 2, 2026 |