Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1930 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Continuous Control | MountainCar Drift II - Dynamics Shift | 18 | Feb 26, 2026 | ||
| Continuous Control | MountainCar Drift I - Dynamics Shift | 18 | Feb 26, 2026 | ||
| Gridworld Navigation | FrozenLake Drift II - Topology Shift | 18 | Feb 26, 2026 | ||
| Gridworld Navigation | FrozenLake Drift I - Topology Shift | 18 | Feb 26, 2026 | ||
| Web Navigation | WebShop Drift II - Semantic Shift | 18 | Feb 26, 2026 | ||
| Web Navigation | WebShop Drift I - Semantic Shift | 18 | Feb 26, 2026 | ||
| Grid-world Navigation | FrozenLake reward reversal Source | 18 | Feb 26, 2026 | ||
| E-commerce Navigation and Search | WebShop semantic shift Source | 18 | Feb 26, 2026 | ||
| Task Efficient Routing | Science scenario | 18 | Feb 26, 2026 | ||
| Code Generation | OJBench ICPC 2025 (test) | 18 | Feb 26, 2026 | ||
| Code Generation | LiveCodeBench lite v6 (test) | 18 | Feb 26, 2026 | ||
| Code Generation | LiveCodeBench lite v5 (test) | 18 | Feb 26, 2026 | ||
| Code Generation | MBPP 2021 (test) | 18 | Apr 14, 2026 | ||
| Online Bin Packing Problem | BPP online N=10k, W=100 | 18 | Apr 8, 2026 | ||
| Multimodal Understanding | ShopBench | 18 | Feb 26, 2026 | ||
| Question Answering | Average 100K context | 18 | Feb 27, 2026 | ||
| Question Answering | Average 30K context | 18 | Feb 27, 2026 | ||
| Question Answering | TriviaQA 100K context | 18 | Feb 27, 2026 | ||
| Question Answering | TriviaQA 30K context | 18 | Feb 27, 2026 | ||
| Question Answering | NQ 100K context | 18 | Feb 27, 2026 | ||
| Question Answering | NQ 30K context | 18 | Feb 27, 2026 | ||
| Question Answering | HotpotQA 100K context | 18 | Feb 27, 2026 | ||
| Question Answering | HotpotQA 30K context | 18 | Feb 27, 2026 | ||
| Mathematical Reasoning | AIME 25 | 18 | Feb 26, 2026 | ||
| Mathematical Reasoning | AIME 24 | 18 | Feb 26, 2026 |