List your company

Research

Benchmarks

Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.

309,131 benchmarksPage 1856 of 12,366

Benchmarks

TaskDatasetDomainTrendResultsLast update
Image ClassificationImageNet-1K Noisy 1.0 (val)computer vision18Apr 9, 2026
Image ClassificationImageNet-1K Clean 1.0 (val)computer vision18Apr 9, 2026
Next POI PredictionCAtabular time series18Apr 9, 2026
Next POI PredictionTKYtabular time series18Apr 9, 2026
Graph GenerationVertex-labeled Synthetic Graphs (test)graphbio medical18Apr 8, 2026
Multi-image Multi-modal Question AnsweringMileBenchmultimodalnlp+118Apr 8, 2026
Auto-biddingAuctionNet Overalltabular time series18May 28, 2026
Auto-biddingAuctionNet P20tabular time series18May 28, 2026
Auto-biddingAuctionNet P19tabular time series18May 28, 2026
Auto-biddingAuctionNet P18tabular time series18May 28, 2026
Auto-biddingAuctionNet P17tabular time series18May 28, 2026
Auto-biddingAuctionNet P16tabular time series18May 28, 2026
Auto-biddingAuctionNet P15tabular time series18May 28, 2026
Auto-biddingAuctionNet P14tabular time series18May 28, 2026
Logical Error Rate (LER) estimationGoogle surface-code data d=3miscellaneous18Apr 8, 2026
Creative WritingBlessing short-textllm chat instruction18Apr 8, 2026
Medical Image SegmentationMSD Prostatebio medicalcomputer vision18May 19, 2026
Mathematical ReasoningGSM8K PROllm chat instructionnlp18Apr 8, 2026
Mathematical ReasoningFOLIO to GSM8Kllm chat instructionnlp18Apr 8, 2026
Visual Document RetrievalViDoRe2 Average (val)retrieval search ragcomputer vision+118Apr 8, 2026
Visual Document RetrievalViDoRe ESG Full 2 (val)retrieval search ragcomputer vision+118Apr 8, 2026
Visual Document RetrievalViDoRe2 ESG Human (val)retrieval search ragcomputer vision+118Apr 8, 2026
Visual Document RetrievalViDoRe2 Econ (val)retrieval search ragcomputer vision+118Apr 8, 2026
Visual Document RetrievalViDoRe Biomed 2 (val)retrieval search ragcomputer vision+118Apr 8, 2026
Score PredictionOmniScore-Bench (test)tabular time series18Apr 8, 2026