Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2280 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Constrained Decoding | SMILES-Bench | 16 | Feb 26, 2026 | ||
| Constrained Decoding | CPP-Bench | 16 | Feb 26, 2026 | ||
| Constrained Decoding | JSON-Bench | 16 | Feb 26, 2026 | ||
| Functional Correctness | CPP-Bench | 16 | Feb 26, 2026 | ||
| Functional Correctness | JSON-Bench | 16 | Feb 26, 2026 | ||
| Response Classification | SEA-SafeguardBench CG Cultural | 16 | Feb 26, 2026 | ||
| Multi-Hop Question Answering | 2WikiMultiHopQA (test val) | 16 | Jul 8, 2026 | ||
| Deep research | HLE | 16 | Feb 26, 2026 | ||
| Long-context understanding | LongEval | 16 | Feb 26, 2026 | ||
| Mathematical Reasoning | MATH500 | 16 | Feb 26, 2026 | ||
| Report Generation | DeepResearch Bench 2025 (test) | 16 | Feb 26, 2026 | ||
| Competitive Coding | CodeForces | 16 | Feb 26, 2026 | ||
| Competitive Coding | ICPC-Eval | 16 | Feb 26, 2026 | ||
| Competitive Coding | OJBench | 16 | Feb 26, 2026 | ||
| Competitive Coding | LiveCodeBench | 16 | Feb 26, 2026 | ||
| Competitive Coding | USACO | 16 | Feb 26, 2026 | ||
| Occupation Classification | Bias-in-Bio lightweight (test) | 16 | Feb 26, 2026 | ||
| Bias Evaluation | BBQ averaged across gender, nationality, and religion domains | 16 | Feb 26, 2026 | ||
| Open Generation | Open Generation | 16 | Feb 26, 2026 | ||
| Long-context language understanding | RULER 4k context length | 16 | May 13, 2026 | ||
| Document Understanding | ChartXiv-DQ | 16 | Feb 26, 2026 | ||
| Embodied Decision-making | ALFWorld (ID and OOD) | 16 | Mar 10, 2026 | ||
| Hate speech classification | Stormfront (test) | 16 | Feb 26, 2026 | ||
| Hate speech classification | Measuring Hate Speech (MHS) (test) | 16 | Feb 26, 2026 | ||
| Hate speech classification | HateCheck (test) | 16 | Feb 26, 2026 |