Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1366 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Mathematical Reasoning | Mathematical Reasoning Benchmarks AIME24, MATH500, Olympiad, AMC, Minerva (test) | 22 | Apr 7, 2026 | ||
| Multimodal Question Answering | Aggregate (Open-WikiTable, 2WikiMQA, InfoSeek, Dyn-VQA, TabFact, WebQA) | 22 | Apr 7, 2026 | ||
| Multimodal Question Answering | WebQA | 22 | Apr 7, 2026 | ||
| Multimodal Question Answering | TabFact | 22 | Apr 7, 2026 | ||
| Multimodal Question Answering | Dyn-VQA | 22 | Apr 7, 2026 | ||
| Multimodal Question Answering | 2WikiMQA | 22 | Apr 7, 2026 | ||
| Multimodal Question Answering | Open-WikiTable | 22 | Apr 7, 2026 | ||
| Adversarial Attack Transferability | ImageNet-Compatible (test) | 22 | Apr 7, 2026 | ||
| Category Unlearning | MNIST (test) | 22 | Apr 7, 2026 | ||
| Input Moderation | WildGuard (test) | 22 | Apr 7, 2026 | ||
| Input Moderation | HarmBench (test) | 22 | Apr 7, 2026 | ||
| Input Moderation | SS (test) | 22 | Apr 7, 2026 | ||
| Input Moderation | AEGIS 2.0 (test) | 22 | Apr 7, 2026 | ||
| Input Moderation | Input Moderation Benchmark Suite (ToxicChat, OAIMod, Aegis, Aegis2, SSTest, HarmB, WildG) | 22 | Apr 7, 2026 | ||
| Relative Pose Estimation | In-Cabin-Pose real-world (test) | 22 | Apr 7, 2026 | ||
| Multimodal reasoning | ZeroBench | 22 | Jun 4, 2026 | ||
| General Knowledge | GPQA Diamond | 22 | Jun 8, 2026 | ||
| Behavior Tree Synthesis | 10 Real-world User Requests 1.0 (test) | 22 | Apr 6, 2026 | ||
| Projector Compensation | Set A same ProCams devices (unseen setups) | 22 | Apr 3, 2026 | ||
| Long-Memory Question Answering | LoCoMo | 22 | May 6, 2026 | ||
| Drone-to-Satellite Retrieval | University-1652 | 22 | Apr 3, 2026 | ||
| Satellite-to-Drone Retrieval | SUES-200 300 m | 22 | Apr 3, 2026 | ||
| Satellite-to-Drone Retrieval | SUES-200 250 m | 22 | Apr 3, 2026 | ||
| Satellite-to-Drone Retrieval | SUES-200 200 m | 22 | Apr 3, 2026 | ||
| Semantic Segmentation | UAVid (val) | 22 | Jul 7, 2026 |