Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 904 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Classification | balance-scale | 29 | May 12, 2026 | ||
| Visual Question Answering | BLINK (val) | 29 | Mar 31, 2026 | ||
| Reasoning | ARC Challenge | 29 | May 13, 2026 | ||
| Oriented Object Detection | DOTA single-scale 1.0 | 29 | Mar 11, 2026 | ||
| Social Reasoning | SimpleToM | 29 | Mar 11, 2026 | ||
| Table Question Answering | WikiTQ | 29 | Mar 11, 2026 | ||
| Long-context language modeling evaluation | RULER | 29 | May 29, 2026 | ||
| Deshadowing | SRD | 29 | Apr 6, 2026 | ||
| Image Styling | Regular Dataset | 29 | Mar 10, 2026 | ||
| Spreadsheet Understanding | FRTR-Bench | 29 | Mar 9, 2026 | ||
| Temporal Action Segmentation | MCFS-130 | 29 | Mar 26, 2026 | ||
| Expert knowledge QA | GPQA | 29 | May 7, 2026 | ||
| Multimodal Evaluation | SEED-Bench 2 Plus | 29 | Mar 9, 2026 | ||
| Object Hallucination Evaluation | POPE GQA (test) | 29 | Apr 1, 2026 | ||
| 5-way classification | Plantae (test) | 29 | Mar 6, 2026 | ||
| 5-way classification | Places (test) | 29 | Mar 6, 2026 | ||
| 5-way classification | Cars (test) | 29 | Mar 6, 2026 | ||
| 5-way classification | CUB (test) | 29 | Mar 6, 2026 | ||
| Retrieval-Augmented Question Answering | RAGQA Leaderboard (test) | 29 | Mar 6, 2026 | ||
| Time-series prediction | ETT | 29 | May 8, 2026 | ||
| Image Aesthetic Assessment | AADB | 29 | Jun 5, 2026 | ||
| Video depth estimation | ScanNet | 29 | Mar 13, 2026 | ||
| Structured Image Editing | StructEditBench | 29 | Mar 5, 2026 | ||
| LLM Agent Evaluation | tau-bench Airline | 29 | May 27, 2026 | ||
| Offline Reinforcement Learning | D4RL Maze2D Medium | 29 | Jun 15, 2026 |