Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1830 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Trajectory-safety classification | ATBench-C | 18 | May 13, 2026 | ||
| Focusable Depth Estimation | RLBench | 18 | May 13, 2026 | ||
| Traffic Prediction | Milan telecommunications INTERNET | 18 | Jun 29, 2026 | ||
| Multi-image retrieval | MS-COCO Random Setting N=4 (test) | 18 | May 13, 2026 | ||
| Video Dehazing | HazeWorld | 18 | May 13, 2026 | ||
| Video Dehazing | UHV-4K | 18 | May 13, 2026 | ||
| Modality Matching | RedditS | 18 | May 13, 2026 | ||
| Modality Matching | Bili_music | 18 | May 13, 2026 | ||
| Modality Matching | Ele-fashion | 18 | May 13, 2026 | ||
| Tool Use | ToolUse | 18 | Jul 3, 2026 | ||
| Code Generation | LiveCodeBench Aug 2024–Feb 2025 v5 | 18 | May 13, 2026 | ||
| Commonsense Reasoning | Commonsense Reasoning Suite (PIQA, SIQA, HellaSwag, ARC-E, ARC-C, WinoGrande, LAMBADA) zero-shot | 18 | May 12, 2026 | ||
| Exploitability Calculation | HUNL Endgame | 18 | May 12, 2026 | ||
| Nash Equilibrium Approximation | HUNL Endgame 10 boards (random boards) | 18 | May 12, 2026 | ||
| Long-form research | ResearchQA | 18 | May 12, 2026 | ||
| Search-based Question Answering | BrowseComp-Plus | 18 | Jun 16, 2026 | ||
| Multimodal Deep Search | HLE-VL | 18 | Jun 9, 2026 | ||
| OOD Detection | OpenOOD Far-OOD | 18 | May 12, 2026 | ||
| OOD Detection | OpenOOD (Near-OOD) | 18 | May 12, 2026 | ||
| Helpful Dialogue | Anthropic HH-RLHF helpful core250 (test) | 18 | May 12, 2026 | ||
| Best-of-N Reward Evaluation | UltraFeedback core250 | 18 | May 12, 2026 | ||
| Reward Modeling | UltraFeedback core250 (held-out evaluation) | 18 | May 12, 2026 | ||
| Model Utility | IFEval | 18 | May 12, 2026 | ||
| Model Utility | Alpaca | 18 | May 12, 2026 | ||
| Jailbreak Defense | Jailbreak Attacks | 18 | May 12, 2026 |