Published benchmarks
Scheduler Latency
A measure of HTTP-based job scheduler latency (how "on time" various job schedulers fire).
Blender Benchmark — CPU
Cycles CPU render performance across community-benchmarked processors.
Blender Benchmark — GPU
Cycles GPU render performance across community-benchmarked graphics devices.
HELM Capabilities
Language models scored on five scenarios by the Stanford HELM Capabilities leaderboard.
Open LLM Leaderboard (archived)
Final scores from the retired Hugging Face Open LLM Leaderboard: open-weight language models on six standardized benchmarks.
ClickBench — analytical databases
Analytical-database runtimes for 43 SQL queries over a ~100M-row web-analytics dataset.
SPEC HPC 2021 — high-performance computing
Audited HPC application-suite results for tested compute systems.
SPECstorage Solution 2020 — storage performance
Audited storage-solution throughput and response time across workload types.
TPC-DS — decision support
Audited decision-support query throughput at a stated data scale factor.
AutoML Benchmark (AMLB)
AutoML frameworks compared on classification accuracy and ROC AUC across a shared set of OpenML tasks.
OpenML-CC18: adult
Best predictive accuracy per machine-learning flow on the adult classification task from the OpenML-CC18 suite.
OpenML-CC18: analcatdata_authorship
Best predictive accuracy per machine-learning flow on the analcatdata_authorship classification task from the OpenML-CC18 suite.
OpenML-CC18: analcatdata_dmft
Best predictive accuracy per machine-learning flow on the analcatdata_dmft classification task from the OpenML-CC18 suite.
OpenML-CC18: balance-scale
Best predictive accuracy per machine-learning flow on the balance-scale classification task from the OpenML-CC18 suite.
OpenML-CC18: bank-marketing
Best predictive accuracy per machine-learning flow on the bank-marketing classification task from the OpenML-CC18 suite.
OpenML-CC18: banknote-authentication
Best predictive accuracy per machine-learning flow on the banknote-authentication classification task from the OpenML-CC18 suite.
OpenML-CC18: Bioresponse
Best predictive accuracy per machine-learning flow on the Bioresponse classification task from the OpenML-CC18 suite.
OpenML-CC18: blood-transfusion-service-center
Best predictive accuracy per machine-learning flow on the blood-transfusion-service-center classification task from the OpenML-CC18 suite.
OpenML-CC18: breast-w
Best predictive accuracy per machine-learning flow on the breast-w classification task from the OpenML-CC18 suite.
OpenML-CC18: car
Best predictive accuracy per machine-learning flow on the car classification task from the OpenML-CC18 suite.
OpenML-CC18: churn
Best predictive accuracy per machine-learning flow on the churn classification task from the OpenML-CC18 suite.
OpenML-CC18: CIFAR_10
Best predictive accuracy per machine-learning flow on the CIFAR_10 classification task from the OpenML-CC18 suite.
OpenML-CC18: climate-model-simulation-crashes
Best predictive accuracy per machine-learning flow on the climate-model-simulation-crashes classification task from the OpenML-CC18 suite.
OpenML-CC18: cmc
Best predictive accuracy per machine-learning flow on the cmc classification task from the OpenML-CC18 suite.
OpenML-CC18: cnae-9
Best predictive accuracy per machine-learning flow on the cnae-9 classification task from the OpenML-CC18 suite.
OpenML-CC18: connect-4
Best predictive accuracy per machine-learning flow on the connect-4 classification task from the OpenML-CC18 suite.
OpenML-CC18: credit-approval
Best predictive accuracy per machine-learning flow on the credit-approval classification task from the OpenML-CC18 suite.
OpenML-CC18: credit-g
Best predictive accuracy per machine-learning flow on the credit-g classification task from the OpenML-CC18 suite.
OpenML-CC18: cylinder-bands
Best predictive accuracy per machine-learning flow on the cylinder-bands classification task from the OpenML-CC18 suite.
OpenML-CC18: Devnagari-Script
Best predictive accuracy per machine-learning flow on the Devnagari-Script classification task from the OpenML-CC18 suite.
OpenML-CC18: diabetes
Best predictive accuracy per machine-learning flow on the diabetes classification task from the OpenML-CC18 suite.
OpenML-CC18: dna
Best predictive accuracy per machine-learning flow on the dna classification task from the OpenML-CC18 suite.
OpenML-CC18: dresses-sales
Best predictive accuracy per machine-learning flow on the dresses-sales classification task from the OpenML-CC18 suite.
OpenML-CC18: electricity
Best predictive accuracy per machine-learning flow on the electricity classification task from the OpenML-CC18 suite.
OpenML-CC18: eucalyptus
Best predictive accuracy per machine-learning flow on the eucalyptus classification task from the OpenML-CC18 suite.
OpenML-CC18: Fashion-MNIST
Best predictive accuracy per machine-learning flow on the Fashion-MNIST classification task from the OpenML-CC18 suite.
OpenML-CC18: first-order-theorem-proving
Best predictive accuracy per machine-learning flow on the first-order-theorem-proving classification task from the OpenML-CC18 suite.
OpenML-CC18: GesturePhaseSegmentationProcessed
Best predictive accuracy per machine-learning flow on the GesturePhaseSegmentationProcessed classification task from the OpenML-CC18 suite.
OpenML-CC18: har
Best predictive accuracy per machine-learning flow on the har classification task from the OpenML-CC18 suite.
OpenML-CC18: ilpd
Best predictive accuracy per machine-learning flow on the ilpd classification task from the OpenML-CC18 suite.
OpenML-CC18: Internet-Advertisements
Best predictive accuracy per machine-learning flow on the Internet-Advertisements classification task from the OpenML-CC18 suite.
OpenML-CC18: isolet
Best predictive accuracy per machine-learning flow on the isolet classification task from the OpenML-CC18 suite.
OpenML-CC18: jm1
Best predictive accuracy per machine-learning flow on the jm1 classification task from the OpenML-CC18 suite.
OpenML-CC18: jungle_chess_2pcs_raw_endgame_complete
Best predictive accuracy per machine-learning flow on the jungle_chess_2pcs_raw_endgame_complete classification task from the OpenML-CC18 suite.
OpenML-CC18: kc1
Best predictive accuracy per machine-learning flow on the kc1 classification task from the OpenML-CC18 suite.
OpenML-CC18: kc2
Best predictive accuracy per machine-learning flow on the kc2 classification task from the OpenML-CC18 suite.
OpenML-CC18: kr-vs-kp
Best predictive accuracy per machine-learning flow on the kr-vs-kp classification task from the OpenML-CC18 suite.
OpenML-CC18: letter
Best predictive accuracy per machine-learning flow on the letter classification task from the OpenML-CC18 suite.
OpenML-CC18: madelon
Best predictive accuracy per machine-learning flow on the madelon classification task from the OpenML-CC18 suite.
OpenML-CC18: mfeat-factors
Best predictive accuracy per machine-learning flow on the mfeat-factors classification task from the OpenML-CC18 suite.
OpenML-CC18: mfeat-fourier
Best predictive accuracy per machine-learning flow on the mfeat-fourier classification task from the OpenML-CC18 suite.
OpenML-CC18: mfeat-karhunen
Best predictive accuracy per machine-learning flow on the mfeat-karhunen classification task from the OpenML-CC18 suite.
OpenML-CC18: mfeat-morphological
Best predictive accuracy per machine-learning flow on the mfeat-morphological classification task from the OpenML-CC18 suite.
OpenML-CC18: mfeat-pixel
Best predictive accuracy per machine-learning flow on the mfeat-pixel classification task from the OpenML-CC18 suite.
OpenML-CC18: mfeat-zernike
Best predictive accuracy per machine-learning flow on the mfeat-zernike classification task from the OpenML-CC18 suite.
OpenML-CC18: MiceProtein
Best predictive accuracy per machine-learning flow on the MiceProtein classification task from the OpenML-CC18 suite.
OpenML-CC18: mnist_784
Best predictive accuracy per machine-learning flow on the mnist_784 classification task from the OpenML-CC18 suite.
OpenML-CC18: nomao
Best predictive accuracy per machine-learning flow on the nomao classification task from the OpenML-CC18 suite.
OpenML-CC18: numerai28.6
Best predictive accuracy per machine-learning flow on the numerai28.6 classification task from the OpenML-CC18 suite.
OpenML-CC18: optdigits
Best predictive accuracy per machine-learning flow on the optdigits classification task from the OpenML-CC18 suite.
OpenML-CC18: ozone-level-8hr
Best predictive accuracy per machine-learning flow on the ozone-level-8hr classification task from the OpenML-CC18 suite.
OpenML-CC18: pc1
Best predictive accuracy per machine-learning flow on the pc1 classification task from the OpenML-CC18 suite.
OpenML-CC18: pc3
Best predictive accuracy per machine-learning flow on the pc3 classification task from the OpenML-CC18 suite.
OpenML-CC18: pc4
Best predictive accuracy per machine-learning flow on the pc4 classification task from the OpenML-CC18 suite.
OpenML-CC18: pendigits
Best predictive accuracy per machine-learning flow on the pendigits classification task from the OpenML-CC18 suite.
OpenML-CC18: PhishingWebsites
Best predictive accuracy per machine-learning flow on the PhishingWebsites classification task from the OpenML-CC18 suite.
OpenML-CC18: phoneme
Best predictive accuracy per machine-learning flow on the phoneme classification task from the OpenML-CC18 suite.
OpenML-CC18: qsar-biodeg
Best predictive accuracy per machine-learning flow on the qsar-biodeg classification task from the OpenML-CC18 suite.
OpenML-CC18: satimage
Best predictive accuracy per machine-learning flow on the satimage classification task from the OpenML-CC18 suite.
OpenML-CC18: segment
Best predictive accuracy per machine-learning flow on the segment classification task from the OpenML-CC18 suite.
OpenML-CC18: semeion
Best predictive accuracy per machine-learning flow on the semeion classification task from the OpenML-CC18 suite.
OpenML-CC18: sick
Best predictive accuracy per machine-learning flow on the sick classification task from the OpenML-CC18 suite.
OpenML-CC18: spambase
Best predictive accuracy per machine-learning flow on the spambase classification task from the OpenML-CC18 suite.
OpenML-CC18: splice
Best predictive accuracy per machine-learning flow on the splice classification task from the OpenML-CC18 suite.
OpenML-CC18: steel-plates-fault
Best predictive accuracy per machine-learning flow on the steel-plates-fault classification task from the OpenML-CC18 suite.
OpenML-CC18: texture
Best predictive accuracy per machine-learning flow on the texture classification task from the OpenML-CC18 suite.
OpenML-CC18: tic-tac-toe
Best predictive accuracy per machine-learning flow on the tic-tac-toe classification task from the OpenML-CC18 suite.
OpenML-CC18: vehicle
Best predictive accuracy per machine-learning flow on the vehicle classification task from the OpenML-CC18 suite.
OpenML-CC18: vowel
Best predictive accuracy per machine-learning flow on the vowel classification task from the OpenML-CC18 suite.
OpenML-CC18: wall-robot-navigation
Best predictive accuracy per machine-learning flow on the wall-robot-navigation classification task from the OpenML-CC18 suite.
OpenML-CC18: wdbc
Best predictive accuracy per machine-learning flow on the wdbc classification task from the OpenML-CC18 suite.
OpenML-CC18: wilt
Best predictive accuracy per machine-learning flow on the wilt classification task from the OpenML-CC18 suite.
TPC-H — decision support
Audited ad-hoc decision-support query throughput at a stated data scale factor.
SPEC CPU 2017 — Floating-Point Rate (SPECrate)
Audited floating-point-throughput results for priced, availability-dated systems.
SPEC CPU 2017 — Floating-Point Speed (SPECspeed)
Audited single-workload floating-point-speed results for priced, availability-dated systems.
SPEC CPU 2017 — Integer Rate (SPECrate)
Audited integer-throughput results for priced, availability-dated systems.
SPEC CPU 2017 — Integer Speed (SPECspeed)
Audited single-workload integer-speed results for priced, availability-dated systems.
SPECjbb 2015 — Java business throughput
Audited Java server throughput (max-jOPS and critical-jOPS) for tested systems.
SPECpower_ssj2008 — server energy efficiency
Audited server power efficiency (overall ssj_ops per watt) for tested systems.
TPC-E — OLTP throughput
Audited brokerage-house OLTP throughput for priced system configurations.
TPC-C — OLTP throughput
Audited on-line transaction-processing throughput for priced system configurations.