List your company

Research

Benchmarks

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

309,131 benchmarksPage 1796 of 12,366

Benchmarks

TaskDatasetDomainTrendResultsLast update
Agent Task ExecutionQWENCLAWBENCHagents tool usellm chat instruction18Jun 16, 2026
AI-generated video detectionGenVidBench 1.0 (test)video generation editingcomputer vision18Jun 16, 2026
Comprehensive Data IntelligenceAggregated Benchmarks (DABench, DSBench, TableBench, DABStep-R)tabular time seriesnlp+318Jun 16, 2026
Data ReasoningDABStep-Rtabular time series18Jun 16, 2026
Table UnderstandingTableBenchnlp18Jun 16, 2026
Data ScienceDSBenchtabular time series18Jun 16, 2026
Function-name selectionBFCL multiple live (300 tasks model)agents tool use18Jun 16, 2026
General ReasoningGSM8K, MATH-500, GPQA-Diamond, TheoremQA Aggregatellm chat instructionnlp18Jun 16, 2026
Theorem-oriented ReasoningTheoremQAnlp18Jun 16, 2026
Scientific ReasoningGPQA Diamondnlp18Jun 16, 2026
Multi-turn Tool-useFTRL, BFCL, and ToolHop Combinedagents tool use18Jun 16, 2026
Multi-turn Tool-useBFCLagents tool use18Jun 16, 2026
Multi-turn Tool-useFTRLagents tool use18Jun 16, 2026
Mathematics ReasoningAIME 2025llm chat instructionnlp18Jun 16, 2026
Speaker VerificationVoxPhrase Eval-2speech audio18Jun 16, 2026
Speaker VerificationVoxPhrase Eval-1speech audio18Jun 16, 2026
Mathematical ReasoningMinervallm chat instructionnlp18Jun 16, 2026
Code VerificationMBPP (test)code developer ai18Jun 16, 2026
Code GenerationHumanEval+ latest (held-out)code developer ai18Jun 16, 2026
Code GenerationLiveCodeBench v5 (held-out)code developer ai18Jun 16, 2026
Mathematical ReasoningMATH-500 latest (held-out)llm chat instructionnlp18Jun 16, 2026
Scanpath GenerationCOCO-Search18 Target Absentcomputer vision18Jun 16, 2026
Biological Relation Identification (EP)BioREDirectbio medicalnlp18Jun 16, 2026
Systematic Review Stopping135 DTA review outcomes (CLEF datasets) (aggregated)nlp18Jun 16, 2026
Textual long-horizon complex reasoningBrowseCompllm chat instruction18Jun 16, 2026