Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2621 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Single-Choice Question Answering | RefMem-Bench | 14 | Jun 2, 2026 | ||
| Multi-Choice Question Answering | RefMem-Bench | 14 | Jun 2, 2026 | ||
| Safety Detection | ATBench-500 | 14 | Jun 2, 2026 | ||
| Local Agentic Search | LocalSearchBench (test) | 14 | Jun 2, 2026 | ||
| Domain Reasoning | DRBench (test) | 14 | Jun 2, 2026 | ||
| Agentic Reasoning | ResearchQA (test) | 14 | Jun 2, 2026 | ||
| Truthfulness | TruthfulQA 4 options (test) | 14 | Jun 2, 2026 | ||
| Mathematical Reasoning | GSM8k 4 options (test) | 14 | Jun 2, 2026 | ||
| Node Classification | Amazon | 14 | Jun 2, 2026 | ||
| Image Denoising | Hill | 14 | Jun 2, 2026 | ||
| Molecular property regression | ZINC12k (test) | 14 | Jun 2, 2026 | ||
| Synthetic Network Generation | YouTube | 14 | Jun 2, 2026 | ||
| Character-Level Language Modeling | text8 (held-out 1M tokens) | 14 | Jun 2, 2026 | ||
| Mispronunciation Detection | CMU Kids (full corpus) | 14 | Jun 2, 2026 | ||
| Agentic Search | GAIA Search | 14 | Jun 2, 2026 | ||
| Agentic Search | xbench DeepSearch | 14 | Jun 2, 2026 | ||
| Agentic Search | BrowseComp-Plus | 14 | Jun 2, 2026 | ||
| Natural Language Understanding | MMLU | 14 | Jun 2, 2026 | ||
| Safety Alignment Evaluation | WildJailbreak (WildJB) | 14 | Jun 2, 2026 | ||
| Safety Alignment Evaluation | Strata | 14 | Jun 2, 2026 | ||
| Safety Alignment Evaluation | StrongReject SR-PAPL | 14 | Jun 2, 2026 | ||
| Safety Alignment Evaluation | StrongReject SR-PAPA | 14 | Jun 2, 2026 | ||
| Safety Alignment Evaluation | StrongReject SR-PAP_M | 14 | Jun 2, 2026 | ||
| Safety Alignment Evaluation | StrongReject SR-Pair | 14 | Jun 2, 2026 | ||
| Safety Alignment Evaluation | StrongReject SR-base | 14 | Jun 2, 2026 |