Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2244 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Mathematical Reasoning | AIME 2025 | 16 | Feb 26, 2026 | ||
| Mathematical Reasoning | AIME 2024 | 16 | Feb 26, 2026 | ||
| Mathematical Reasoning | GSM8K | 16 | Feb 26, 2026 | ||
| Mathematical Problem Solving | AIME VeRA-H Pro 2024-II | 16 | Feb 26, 2026 | ||
| Mathematical Problem Solving | AIME VeRA-H 2024-II | 16 | Feb 26, 2026 | ||
| Mathematical Problem Solving | AIME Seeds 2024 II | 16 | Feb 26, 2026 | ||
| Text Generation | Decoding Throughput | 16 | Feb 26, 2026 | ||
| Mathematical reasoning | MATH | 16 | Jun 1, 2026 | ||
| Lightning Prediction | Lightning Dataset 6 hours cumulative (test) | 16 | Feb 26, 2026 | ||
| Lightning Prediction | Lightning Dataset 3 hours cumulative (test) | 16 | Feb 26, 2026 | ||
| Lightning Prediction | Lightning Dataset 1 hour cumulative (test) | 16 | Feb 26, 2026 | ||
| Node Classification | CS Homophilic (10 train val test splits) | 16 | Feb 26, 2026 | ||
| Node Classification | Physics Homophilic (10 train/val/test) | 16 | Jun 1, 2026 | ||
| LLM Unlearning | NEWS | 16 | Feb 26, 2026 | ||
| LLM Unlearning | BOOKS | 16 | Feb 26, 2026 | ||
| Mathematical Reasoning | Mathematical Reasoning Benchmarks (test) | 16 | Feb 26, 2026 | ||
| Web Navigation and Automation | WorkArena Held-out Tasks (test) | 16 | Feb 26, 2026 | ||
| Web Navigation and Automation | WorkArena Held-out Goals (test) | 16 | Feb 26, 2026 | ||
| Web Task Completion | MiniWoB++ Held-out Tasks (test) | 16 | Feb 26, 2026 | ||
| Web Task Completion | MiniWoB++ Held-out Goals (test) | 16 | Feb 26, 2026 | ||
| Pairwise LLM Judging | Chatbot Arena | 16 | Feb 26, 2026 | ||
| Pairwise LLM Judging | RewardBench | 16 | Feb 26, 2026 | ||
| Pairwise LLM Judging | MT-Bench | 16 | Jun 1, 2026 | ||
| Perception | TWI-oriented TreeBench online setting | 16 | Feb 26, 2026 | ||
| Reasoning | TWI-oriented TreeBench online setting | 16 | Feb 26, 2026 |