Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2381 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Massive Multitask Language Understanding | MMLU | 15 | May 28, 2026 | ||
| Instruction Following | IFEval | 15 | May 28, 2026 | ||
| Mathematical Reasoning | GSM8K | 15 | May 28, 2026 | ||
| Framing Sensitivity Evaluation | FRAGILE Overall | 15 | May 28, 2026 | ||
| Framing Sensitivity Evaluation | FRAGILE Narrative vividness | 15 | May 28, 2026 | ||
| Framing Sensitivity Evaluation | FRAGILE Temporal slice | 15 | May 28, 2026 | ||
| Framing Sensitivity Evaluation | FRAGILE Value-tinted narration | 15 | May 28, 2026 | ||
| Scene-classification | EuroSAT-MS | 15 | May 28, 2026 | ||
| Legal Judgment Prediction | CAIL & CMDL Average | 15 | May 28, 2026 | ||
| Legal Judgment Prediction | CMDL | 15 | May 28, 2026 | ||
| Legal Judgment Prediction | CAIL 2018 | 15 | May 28, 2026 | ||
| Reaction Diagram Parsing | RxnScribe | 15 | May 28, 2026 | ||
| Subjective Performance Evaluation | Subjective Benchmarks | 15 | May 28, 2026 | ||
| Objective Performance Evaluation | Objective Benchmarks | 15 | May 28, 2026 | ||
| Next-move prediction | M1 | 15 | May 28, 2026 | ||
| Code Generation | PCS (test) | 15 | May 28, 2026 | ||
| Image Classification | CIFAR-10N | 15 | May 28, 2026 | ||
| Object Pointing | VisDrone | 15 | Jul 8, 2026 | ||
| Rule-level anomaly detection | CLEVRER | 15 | May 27, 2026 | ||
| Camera-to-gravity rotation estimation | Hypersim (held out scenes) | 15 | Jun 4, 2026 | ||
| Chart Question Answering | ChartMuseum | 15 | May 27, 2026 | ||
| Page Retrieval | MMDocIR 1.0 (full) | 15 | May 27, 2026 | ||
| Text-to-image generation | FLUX.1 200 prompts (dev) | 15 | May 27, 2026 | ||
| Tax Evasion Detection | T15S | 15 | May 27, 2026 | ||
| Point Cloud Denoising | PCNet 50K points, 3% Gaussian noise | 15 | May 27, 2026 |