Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1562 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Fair Clustering | SYNTHETIC-A | 20 | May 14, 2026 | ||
| Fair Clustering | SYNTHETIC-B | 20 | May 14, 2026 | ||
| Multi-Agent Path Finding | Large Maze 33x33 world size, 33.0% static obstacle rate | 20 | May 14, 2026 | ||
| Multi-Agent Path Finding | Medium Warehouse 25x25 world size, 34.6% static obstacle rate | 20 | May 14, 2026 | ||
| Multi-Agent Path Finding | Medium Room 23x23 world size, 33.5% static obstacle rate | 20 | May 14, 2026 | ||
| Multi-Agent Path Finding | Medium Maze 25x25 world size, 32.8% static obstacle rate | 20 | May 14, 2026 | ||
| Multi-Agent Path Finding | Small Random 10x10 world size, 17.5% static obstacle rate | 20 | May 14, 2026 | ||
| 3D Shape Registration | DT4D-H (test) | 20 | May 14, 2026 | ||
| 3D Shape Registration | SHREC07-H (test) | 20 | May 14, 2026 | ||
| Hallucination Detection | HallusionBench | 20 | Jun 1, 2026 | ||
| Image Classification | FusionBench 20-task suite | 20 | May 14, 2026 | ||
| Image Classification | FusionBench 14-task suite | 20 | May 14, 2026 | ||
| Centerline Tracking | Barcelona-Catalunya GP circuit | 20 | May 14, 2026 | ||
| Tabular multi-class classification | HardCOp 38 datasets Original | 20 | May 14, 2026 | ||
| Time Series Forecasting | Traffic 5% train (test) | 20 | May 14, 2026 | ||
| Time Series Forecasting | Electricity 5% train data (test) | 20 | May 14, 2026 | ||
| Time Series Forecasting | ETTm1 5% train (test) | 20 | May 14, 2026 | ||
| Time Series Forecasting | ETTh1 5% training data (test) | 20 | May 14, 2026 | ||
| Time Series Forecasting | ETTm1 10% (train) | 20 | May 14, 2026 | ||
| Natural Language Understanding and Reasoning | LLM Downstream Evaluation Suite (ARC-c, BoolQ, Winogrande, GSM8k, MMLU) zero-shot | 20 | May 14, 2026 | ||
| Open-ended MCQA | MMLU 16 subjects filtered | 20 | May 14, 2026 | ||
| Tool-use reasoning | ToolAlpaca | 20 | May 26, 2026 | ||
| Top-K Selection | Array of 262,144 elements Top-1024 selection (test) | 20 | May 14, 2026 | ||
| Empirical coverage estimation | RiverSwim T=50 90% nominal coverage | 20 | May 13, 2026 | ||
| Optimal Policy Recovery (Empirical Coverage) | RiverSwim T=50 nominal 95% coverage | 20 | May 13, 2026 |