Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1747 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Language Modelling | Explanatory sentences | 19 | Feb 26, 2026 | ||
| Language Modelling | Mathematical expression EVAL (test) | 19 | Feb 26, 2026 | ||
| Multi-modal Preference Evaluation | MM-RewardBench | 19 | Feb 26, 2026 | ||
| Multi-modal Preference Evaluation | VL-Reward | 19 | Feb 26, 2026 | ||
| Long-CoT Question Generation | CoDiQ-Bench 1.0 (test) | 19 | Feb 26, 2026 | ||
| Coding | LiveCodeBench | 19 | Mar 4, 2026 | ||
| Aggregated Performance | Average 10 Tasks | 19 | Feb 27, 2026 | ||
| Retrieval | PopQA | 19 | Apr 14, 2026 | ||
| Retrieval | NQ | 19 | Apr 14, 2026 | ||
| Reasoning and Math Problem Solving | BoolQ, COPA, WinoG., SVAMP, MAWPS (test) | 19 | Feb 26, 2026 | ||
| Multimodal Reasoning | MathVerse Vision Only | 19 | May 22, 2026 | ||
| Story generation | Story | 19 | May 12, 2026 | ||
| Code Infilling | HumanEval Infilling | 19 | Jun 12, 2026 | ||
| Mathematical Reasoning | GSM8K-NL (test) | 19 | Feb 26, 2026 | ||
| Computer-use Task Execution | All Environments Overall | 19 | Feb 26, 2026 | ||
| Computer-use Task Execution | Celestia | 19 | Feb 26, 2026 | ||
| Computer-use Task Execution | KAlgebra | 19 | Feb 26, 2026 | ||
| Computer-use Task Execution | Thunderbird | 19 | Feb 26, 2026 | ||
| Computer-use Task Execution | Writer | 19 | Feb 26, 2026 | ||
| Computer-use Task Execution | Calc | 19 | Feb 26, 2026 | ||
| Computer-use Task Execution | Impress | 19 | Feb 26, 2026 | ||
| Mathematical reasoning | MATH500 0-shot | 19 | Feb 26, 2026 | ||
| Mathematical reasoning | GSM8K 0-shot | 19 | Feb 26, 2026 | ||
| Reasoning | Sudoku (test) | 19 | May 13, 2026 | ||
| Traveling Salesman Problem | TSP N=50 (test) | 19 | Feb 26, 2026 |