Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1990 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Agentic Terminal Interaction | Terminal-Bench v1 | 17 | Jun 17, 2026 | ||
| Coreference Resolution | CIN (test) | 17 | Jun 17, 2026 | ||
| Graph Classification | MoleculeNet | 17 | Jun 17, 2026 | ||
| Binary Outcome Discrimination | Resolved vs. Rest (test) | 17 | Jun 17, 2026 | ||
| Inpaint (box) | ImageNet 256x256 | 17 | Jun 30, 2026 | ||
| Multi-Task Recommendation | Shopee (offline) | 17 | Jun 16, 2026 | ||
| GPU kernel optimization | KernelBench Level 3 | 17 | Jun 16, 2026 | ||
| GPU kernel optimization | KernelBench Level 1 | 17 | Jun 16, 2026 | ||
| Image Classification | MTIL Transfer (test) | 17 | Jun 16, 2026 | ||
| Video Understanding | CL-VISTA (test) | 17 | Jun 16, 2026 | ||
| Visual Grounding | V* | 17 | Jun 16, 2026 | ||
| Visual Grounding | HR-Bench 4K | 17 | Jun 16, 2026 | ||
| Textual long-horizon complex reasoning | BrowseComp ZH | 17 | Jun 16, 2026 | ||
| Remote code execution (RCE) exploit generation | RCE (test) | 17 | Jun 16, 2026 | ||
| Offline Multi-Objective Optimization | Off-MOO-Bench Sci-Design (50th percentile) | 17 | Jun 16, 2026 | ||
| Offline Multi-Objective Optimization | Off-MOO-Bench MO-NAS (50th percentile) | 17 | Jun 16, 2026 | ||
| Fuzzy Private Set Intersection | L1-FPSI Set Size m=n=2^12, Dimension d=4, WAN | 17 | Jun 16, 2026 | ||
| Fuzzy Private Set Intersection | Synthetic Set Size m=n=2^{12}, Dimension d=4 | 17 | Jun 16, 2026 | ||
| MRI reconstruction | COR-PD (test) | 17 | Jun 16, 2026 | ||
| Meme Classification | Bn-HIB 1.0 (test) | 17 | Jun 16, 2026 | ||
| Odometry | WHU-Helmet | 17 | Jun 16, 2026 | ||
| Dynamic Path Planning | Dynamic Scenarios Sparse Map Type | 17 | Jun 15, 2026 | ||
| Dynamic Path Planning | Dynamic Scenarios Short Map Type | 17 | Jun 15, 2026 | ||
| 3D Point Cloud Classification | ShapeNet-C | 17 | Jun 12, 2026 | ||
| Natural Language Generation | NLG Evaluation Suite (GSM8K, MATH, HumanEval, MBPP, Commonsense) | 17 | Jun 12, 2026 |