Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1550 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Machine Translation | News Commentary Xx ⇒ En V18.1 | 20 | May 29, 2026 | ||
| Machine Translation | News Commentary En ⇒ Xx V18.1 | 20 | May 29, 2026 | ||
| Serving Latency | MATH500, GSM8K, SharedGPT, and HumanEval (test) | 20 | May 29, 2026 | ||
| Image Classification | MNIST | 20 | Jun 2, 2026 | ||
| Natural language inference | MNLI GLUE (test) | 20 | May 29, 2026 | ||
| Topic-oriented text understanding | Yahoo StandardCL (test) | 20 | May 29, 2026 | ||
| Review understanding | Amazon StandardCL (test) | 20 | May 29, 2026 | ||
| Review understanding | Yelp StandardCL (test) | 20 | May 29, 2026 | ||
| Kernel Regression | Boston 20% n=506 (test) | 20 | May 29, 2026 | ||
| Video Moment Retrieval | ActivityNet Captions | 20 | Jun 2, 2026 | ||
| Code Generation | HumanEval | 20 | May 29, 2026 | ||
| Mathematical Reasoning | AIME 25 | 20 | May 29, 2026 | ||
| AI-Generated Image Detection | DRCT-2M | 20 | May 29, 2026 | ||
| Language Modeling | WT-103 | 20 | May 29, 2026 | ||
| Multi-step Reasoning (Math & Code) | Multi-step Reasoning Suite GSM8K MBPP HumanEval | 20 | May 29, 2026 | ||
| General Language Understanding | General Ability Suite (MMLU, PIQA, ARC-E, ARC-C, BoolQ, WinoGrande, HellaSwag, TruthfulQA) | 20 | May 29, 2026 | ||
| Math Reasoning | MATH 500 | 20 | May 29, 2026 | ||
| Function Calling | API-Bank | 20 | Jun 16, 2026 | ||
| 3D Surface Reconstruction | Custom Near-Range mmWave Radar Dataset | 20 | May 29, 2026 | ||
| Score Estimation | 2D Student-t distribution nu=3 | 20 | May 29, 2026 | ||
| False Data Injection Attack (FDIA) Detection | IEEE 118-bus system | 20 | May 29, 2026 | ||
| False Data Injection Attack (FDIA) Detection | IEEE 57-bus system | 20 | May 29, 2026 | ||
| False Data Injection Attack (FDIA) Detection | IEEE 30-bus system | 20 | May 29, 2026 | ||
| False Data Injection Attack (FDIA) Detection | IEEE 14-bus system | 20 | May 29, 2026 | ||
| Token-level Uncertainty Quantification | Generated-DI | 20 | May 28, 2026 |