Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1391 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Mathematical Reasoning | GSM8K | 22 | Jun 30, 2026 | ||
| Continual routing | Average | 22 | Feb 26, 2026 | ||
| Continual routing | GSM8k | 22 | Feb 26, 2026 | ||
| Continual routing | MMLU | 22 | Feb 26, 2026 | ||
| Continual routing | 2WikiMultiHop | 22 | Feb 26, 2026 | ||
| Role-playing Reward Modeling | RoleRM-Bench | 22 | Feb 26, 2026 | ||
| Multiple Choice STEM Question Answering | OlmoBaseEval MCSTEM | 22 | Apr 7, 2026 | ||
| Latent multi-hop reasoning | 2WikiMultiHopQA | 22 | Feb 26, 2026 | ||
| Latent multi-hop reasoning | HotpotQA | 22 | Feb 26, 2026 | ||
| Mathematical Reasoning | AIME 2025 | 22 | Feb 26, 2026 | ||
| Acoustic Discriminability (ABX) | Zero Resource Speech Challenge (en, fr, zh, de, wo) 2017 | 22 | Feb 26, 2026 | ||
| Acoustic Discriminability (ABX) | 5 Languages (sw, ta, th, tr, uk) (dev) | 22 | Feb 26, 2026 | ||
| Embodied Interaction | ALFWorld | 22 | Apr 6, 2026 | ||
| Reasoning | HMMT Feb. 2025 | 22 | Jul 7, 2026 | ||
| Machine Translation (other-language-to-English) | Neko (test) | 22 | Feb 26, 2026 | ||
| Safety Evaluation | Jailbreak Bench | 22 | Apr 2, 2026 | ||
| Community Alignment | Community Alignment (CA) | 22 | Feb 26, 2026 | ||
| Cultural Alignment | WorldValuesBench (WVB) | 22 | Feb 26, 2026 | ||
| Mathematical & Scientific Reasoning | GPQA | 22 | May 14, 2026 | ||
| Flight Recommendation | Flight Rec. | 22 | Feb 26, 2026 | ||
| Mixed 20 Question | Mixed 20Q | 22 | Feb 26, 2026 | ||
| Personalization | LaMP-2 | 22 | Mar 4, 2026 | ||
| Competitive Programming | LiveCodeBench v5 | 22 | Feb 26, 2026 | ||
| Reward Modeling | WPB | 22 | Feb 26, 2026 | ||
| Reward Modeling | WQ Arena | 22 | Feb 26, 2026 |