Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1631 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| TP | Habitat | 20 | Feb 26, 2026 | ||
| Audio-Visual Event Classification | AVE | 20 | Jun 11, 2026 | ||
| Image-to-Image Translation (Appearance Divergence) | LAION Mini | 20 | Feb 26, 2026 | ||
| Image-to-Image Translation (Appearance Consistency) | LAION Mini | 20 | Feb 26, 2026 | ||
| Zero-shot Classification | ImageNet | 20 | May 19, 2026 | ||
| Image Classification | General-ImageNet | 20 | Feb 27, 2026 | ||
| Image Quality Assessment | JND Dataset | 20 | Feb 26, 2026 | ||
| Multimodal Understanding | MMMU Pro | 20 | Feb 26, 2026 | ||
| Declouding | RS-Cloud | 20 | Jun 18, 2026 | ||
| Sim-to-real adaptation | DoTA (target) | 20 | Feb 26, 2026 | ||
| Sub-workflow Execution | OS-Marathon Expense Report - Spreadsheet Environment | 20 | Feb 26, 2026 | ||
| Sub-workflow Execution | OS-Marathon Expense Report - Website System Environment | 20 | Feb 26, 2026 | ||
| GUI Task Automation | OS-Marathon Transcript domain (Level 2) | 20 | Feb 26, 2026 | ||
| Person Re-identification | AG-ReID Protocol 2: G↔A v1 | 20 | Feb 26, 2026 | ||
| Person Re-identification | AG-ReID Protocol 1: A↔G v1 | 20 | Feb 26, 2026 | ||
| Multimodal Search-based Question Answering | MMSearch+ | 20 | Feb 26, 2026 | ||
| Visual Question Answering | VDR | 20 | Feb 26, 2026 | ||
| Synthetic Image Attribution | AbC Artchive Object | 20 | Feb 26, 2026 | ||
| Synthetic Image Attribution | AbC Artchive, GPT | 20 | Feb 26, 2026 | ||
| Synthetic Image Attribution | AbC ImageNet-Unseen Media | 20 | Feb 26, 2026 | ||
| Synthetic Image Attribution | AbC ImageNet-Unseen GPT | 20 | Feb 26, 2026 | ||
| Synthetic Image Attribution | AbC BAM-FG Object | 20 | Feb 26, 2026 | ||
| Synthetic Image Attribution | AbC BAM-FG GPT | 20 | Feb 26, 2026 | ||
| Synthetic Image Attribution | AbC ImageNet-Seen Media | 20 | Feb 26, 2026 | ||
| Synthetic Image Attribution | AbC ImageNet-Seen GPT | 20 | Feb 26, 2026 |