List your company

Research

Benchmarks

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

309,131 benchmarksPage 1525 of 12,366

Benchmarks

TaskDatasetDomainTrendResultsLast update
Brain Disorder ClassificationUCLA (Tenfold cross-validation)bio medical21Apr 21, 2026
Conversational PerformanceTG-REDIAL (test)llm chat instructionnlp21Feb 26, 2026
Zero-shot tasksLM Evaluation Harness Suite zero-shot v1llm chat instructionnlp21Jul 2, 2026
LLM RoutingCEVAL and GSM8K (OOD)llm chat instructionnlp21Feb 26, 2026
LLM RoutingMMLU, CMMLU, etc. In-distributionllm chat instructionnlp21Feb 26, 2026
Multi-Object TrackingWaymo Open Dataset (WOD) Level 2 (test)computer vision21Feb 26, 2026
Multi-Object TrackingWaymo Open Dataset (WOD) Level 1 (test)computer vision21Feb 26, 2026
Video CaptioningMCTS-VCBmultimodal21Feb 26, 2026
Novel View SynthesisDeep Blending3d vision geometry21Feb 26, 2026
Fine-grained Visual ReasoningV* OODcomputer visionmultimodal20Jul 9, 2026
Fine-grained Visual ReasoningHR-Bench-4K OODcomputer visionmultimodal20Jul 9, 2026
Fine-grained Visual ReasoningMME-RW-Lite IDcomputer visionmultimodal20Jul 9, 2026
Ultrasound Image SegmentationCAMUScomputer visionbio medical20Jul 9, 2026
Engineering ReasoningMMLU-Pro Engineeringllm chat instruction20Jul 9, 2026
Physics ReasoningMMLU-Pro Physicsmiscellaneous20Jul 9, 2026
Image ClassificationCIFAR-100computer vision20Jul 9, 2026
Heart Rate MeasurementHRM Prototype Experimental Datasetcomputer visionbio medical20Jul 9, 2026
Nodule & Tumor SegmentationNodule & Tumor Averagecomputer visionbio medical20Jul 8, 2026
Skin Lesion SegmentationSkin Lesion Averagecomputer visionbio medical20Jul 8, 2026
Inverse KinematicsFetch Arm open-world random target (test)robotics embodied ai3d vision geometry20Jul 8, 2026
Inverse KinematicsPanda Arm open-world random target (test)robotics embodied ai3d vision geometry20Jul 8, 2026
Binary ClassificationRelbench v1tabular time series20Jul 8, 2026
Watermark DecodingLlama2-7B Tokenizer 400k tokens (test)computer vision20Jul 7, 2026
Euclidean estimation errorMonte Carlo simulation n = 3 × 10^8 (final-horizon)miscellaneous20Jul 7, 2026
Scientific Problem SolvingPhysicsQAmiscellaneous20Jul 7, 2026