List your company

Research

Benchmarks

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

309,131 benchmarksPage 1947 of 12,366

Benchmarks

TaskDatasetDomainTrendResultsLast update
DialogueDialoguellm chat instructionnlp18Feb 26, 2026
Image ClassificationImageNet 1k (val)computer vision18Feb 26, 2026
Multi-object trackingUAVDT (test)computer vision18Feb 26, 2026
3D Object RecognitionModelNet40 1k points (test)3d vision geometry18Feb 26, 2026
Object Hallucination EvaluationPOPE 28multimodalcomputer vision18Feb 26, 2026
Multimodal Perception EvaluationMME Perception 16multimodalcomputer vision18Feb 26, 2026
Novel View SynthesisNuscenes3d vision geometry18May 19, 2026
Direct-answer Visual Question AnsweringA-OKVQAmultimodalcomputer vision+118Feb 26, 2026
AI-generated text detectionHC3 (test)nlp18Feb 26, 2026
Object Hallucination EvaluationMME Existence (test)multimodalcomputer vision18Feb 26, 2026
Visual PerceptionSEED-Bench Imagecomputer visionmultimodal18Feb 26, 2026
Knowledge Base Question AnsweringCWQnlpretrieval search rag18Feb 26, 2026
Dog Whistle Presence DetectionSynthetic-Detection 1.0 (test)computer vision18Feb 26, 2026
Entity LocalizationVRDU Registration Form (test)computer vision18Feb 27, 2026
Remote Sensing Change DetectionSYSU-CD (test)computer vision18Feb 26, 2026
Machine TranslationWMT 22 (test)nlp18Feb 26, 2026
Machine TranslationWMT En-Zh 22 (test)nlp18Feb 26, 2026
DisentanglementARG0computer visionbio medical+118Feb 26, 2026
Knowledge Base Question AnsweringGrailQA 500-sample (dev)nlpretrieval search rag18Feb 26, 2026
Simultaneous Speech TranslationCallHome Spanish-English Es-En (test)speech audionlp18Feb 26, 2026
Code GenerationMBPP subset of 500 (test)code developer ai18Feb 26, 2026
Long-Term AnticipationEgo4D LTA v1 (test)robotics embodied aicomputer vision18Feb 26, 2026
Adversarial Attack DefenseGCG Individualadversarial safety security18Feb 26, 2026
Fashion video synthesisUBC fashion video dataset (test)video generation editing18Apr 1, 2026
Instruction FollowingVicuna benchmarkllm chat instruction18Feb 26, 2026