Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2283 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Multi-hop Question Answering | HotpotQA 800 documents | 16 | Feb 26, 2026 | ||
| Multi-hop Question Answering | HotpotQA 400 documents | 16 | Feb 26, 2026 | ||
| Question Answering | Average (PopQA, HotpotQA, 2Wiki, MuSiQue, Bamboogle) | 16 | Jul 9, 2026 | ||
| End-to-end Tool-use | ToolBench I1 v1 | 16 | Feb 26, 2026 | ||
| Reasoning | AIME 25 | 16 | Apr 23, 2026 | ||
| Mathematical Reasoning | OlympiadBench (OLY.) | 16 | Feb 26, 2026 | ||
| Hallucination | HallusionBench | 16 | Feb 26, 2026 | ||
| Interactive Agent Task | TRIP-Bench Overall | 16 | Feb 26, 2026 | ||
| Interactive Agent Task | TRIP-Bench Hard PMR | 16 | Feb 26, 2026 | ||
| Interactive Agent Task | TRIP-Bench (Hard FIT) | 16 | Feb 26, 2026 | ||
| Interactive Agent Task | TRIP-Bench Mid | 16 | Feb 26, 2026 | ||
| Interactive Agent Task | TRIP-Bench Easy | 16 | Feb 26, 2026 | ||
| Deep search | hle | 16 | Jul 1, 2026 | ||
| Regression | Parkinsons | 16 | Feb 26, 2026 | ||
| GUI Agent | AC High | 16 | Feb 26, 2026 | ||
| GUI Agent | AC Low | 16 | Feb 26, 2026 | ||
| Reward Modeling | MMRB 2 | 16 | Jun 26, 2026 | ||
| Mathematical Reasoning | Verify-RL (D5) | 16 | Feb 26, 2026 | ||
| Concept Unlearning | MMA-Diffusion | 16 | Feb 26, 2026 | ||
| Medical LLM Evaluation | LLMEval Med | 16 | Mar 23, 2026 | ||
| Medical Question Answering | HealthBench Overall | 16 | Feb 26, 2026 | ||
| Open-ended text generation | Law-MT Out of Domain (test) | 16 | Feb 26, 2026 | ||
| Instruction Following | IFEval | 16 | Apr 3, 2026 | ||
| Length Prediction | ForeLen Avg. 1.0 (test) | 16 | Feb 26, 2026 | ||
| Length Prediction | ForeLen RL 1.0 (test) | 16 | Feb 26, 2026 |