Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1432 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Knowledge cutoff success | Semantic | 21 | Jun 5, 2026 | ||
| Knowledge cutoff success | Factual | 21 | Jun 5, 2026 | ||
| Audio-Visual Speech Recognition | LRS3 Clean original (test) | 21 | Jun 5, 2026 | ||
| Input-guided generation | Spec-Bench | 21 | Jun 5, 2026 | ||
| Reasoning | MMLU-Pro | 21 | Jun 5, 2026 | ||
| Reasoning | MATH 500 | 21 | Jun 5, 2026 | ||
| Reasoning | AIME 2024 | 21 | Jun 5, 2026 | ||
| Image Reconstruction | ImageNet 256 | 21 | Jun 12, 2026 | ||
| Language Modeling | WikiText-103 Attacked (test) | 21 | Jun 18, 2026 | ||
| Language Modeling | WikiText-103 Attacked (val) | 21 | Jun 18, 2026 | ||
| Agentic decision making | Movie | 21 | Jun 5, 2026 | ||
| Agentic decision making | Weather | 21 | Jun 5, 2026 | ||
| Reasoning | ARC-E, CSQA, Lambada, PIQA, Wino | 21 | Jun 5, 2026 | ||
| Adversarial Attack | SVHN D | 21 | Jun 5, 2026 | ||
| Kernel Ridge Regression | CPU (test) | 21 | Jun 5, 2026 | ||
| Language Model Evaluation | Aggregate | 21 | Jun 4, 2026 | ||
| Person understanding and persistent memory | LoCoMo-Plus | 21 | Jun 4, 2026 | ||
| Person understanding and persistent memory | RealMem | 21 | Jun 4, 2026 | ||
| Reasoning | Out-of-Domain (OOD) Reasoning Suite (ARC-c, GPQA*, MMLU-Pro) various (test) | 21 | Jun 4, 2026 | ||
| LLM Decoding Throughput | NOSA-8B 64K context length | 21 | Jun 4, 2026 | ||
| LLM Decoding Throughput | MiniCPM 8B 64K context length 4.1 | 21 | Jun 4, 2026 | ||
| Single-turn Jailbreak Attack | JailbreakBench single-turn | 21 | Jun 4, 2026 | ||
| Face Recognition Probability Calibration | BFW | 21 | Jun 4, 2026 | ||
| Face Recognition Probability Calibration | RFW | 21 | Jun 4, 2026 | ||
| Medical Image Editing | Med-Banana-80K v1.0 (test) | 21 | Jun 4, 2026 |