Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2130 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Route Tracing | MapTrace | 16 | Jun 12, 2026 | ||
| Visual Probing | VisualProbe Hard | 16 | Jun 12, 2026 | ||
| Visual Probing | VisualProbe Medium | 16 | Jun 12, 2026 | ||
| Visual Probing | VisualProbe Easy | 16 | Jun 12, 2026 | ||
| Spatial Reasoning | ReasonMap | 16 | Jun 12, 2026 | ||
| Spatial Reasoning | ReasonMap Plus | 16 | Jun 12, 2026 | ||
| Multicalibration | COMPAS | 16 | Jun 12, 2026 | ||
| Certified Robustness Verification | MNIST first 50 samples (test) | 16 | Jun 12, 2026 | ||
| Tool Attack Defense | Multi-Agent System (MAS) Topologies Average of Chain, Tree, Star, Random | 16 | Jun 12, 2026 | ||
| Prompt Injection Defense | Multi-Agent System (MAS) Topologies Average of Chain, Tree, Star, Random | 16 | Jun 12, 2026 | ||
| Robot Manipulation | ManiSkill Table-Top 2 Finger Gripper (Overall) | 16 | Jun 11, 2026 | ||
| Robot Manipulation | ManiSkill RollBall v1 | 16 | Jun 11, 2026 | ||
| Robot Manipulation | ManiSkill PokeCube v1 | 16 | Jun 11, 2026 | ||
| Robot Manipulation | ManiSkill PushCube v1 | 16 | Jun 11, 2026 | ||
| Short-video Question Answering | Short-video QA Suite NextQA, PerceptionTest, MVBench, Tomato, MotionBench, TempCompass | 16 | Jun 11, 2026 | ||
| SCPI extraction | SCPI extraction corpus | 16 | Jun 11, 2026 | ||
| Locomotion | HalfCheetah | 16 | Jun 11, 2026 | ||
| Robotic Manipulation | Meta-World modified (test) | 16 | Jun 11, 2026 | ||
| Theory of Mind reasoning | Hi-ToM | 16 | Jun 11, 2026 | ||
| Theory of Mind Reasoning | FanToM | 16 | Jun 11, 2026 | ||
| Agentic Performance | OSWorld Verified | 16 | Jun 11, 2026 | ||
| Concept Faithfulness Evaluation | AGNEWS | 16 | Jun 11, 2026 | ||
| Concept Faithfulness Evaluation | Jigsaw | 16 | Jun 11, 2026 | ||
| Concept Faithfulness Evaluation | ERASER Movie | 16 | Jun 11, 2026 | ||
| Aggregated Multi-task Evaluation | Average | 16 | Jun 11, 2026 |