Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Performance Prediction research benchmarks · SOTA2 Research

Performance Prediction

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Small CNN Zoo ReLU subset (test)

0.926Kendall’s Tau

35

Apr 28, 2026

Reasoning Benchmarks Average

0.874Train R^2

21

Jul 2, 2026

WMT Benchmarks Average WMT'14 & WMT'19 (aggregation)

0.29MAE

16

Feb 26, 2026

WMT En-De 2019 (val)

0.29MAE

16

Feb 26, 2026

WMT En-Fr 2014 (val)

0.28MAE

16

Feb 26, 2026

WMT En-De 2014 (val)

0.22MAE

16

Feb 26, 2026

NAS-Bench-101 (val)

94.76Accuracy

11

May 12, 2026

ARC 1.2k (test)

1.14MAE

11

Mar 4, 2026

Winogrande (WG) 1.3k (test)

1MAE

11

Mar 4, 2026

HellaSwag (HS) 10k (test)

0.8MAE

11

Mar 4, 2026

MMLU 14k (test)

1.07MAE

11

Mar 4, 2026

Large Model Performance Prediction Dataset 80% masking (test)

7.5RMSE

10

Feb 26, 2026

MNLI source domains (out-of-domain)

Cosine distance (fine-tuned)

0.683ROC AUC

10

Feb 26, 2026

MNLI source domains (in-domain)

Cosine distance (fine-tuned)

0.699ROC AUC

10

Feb 26, 2026

Sentiment temporal (out-of-domain)

Cosine distance (fine-tuned)

0.834ROC AUC

10

Feb 26, 2026

Sentiment temporal (in-domain)

Cosine distance (fine-tuned)

0.852ROC AUC

10

Feb 26, 2026

Sentiment categories (out-of-domain)

Cosine distance (fine-tuned)

0.822ROC AUC

10

Feb 26, 2026

Sentiment categories (in-domain)

Cosine distance (fine-tuned)

0.845ROC AUC

10

Feb 26, 2026

NAS-Bench-201 ImageNet16 (test)

0.63Kendall's Tau

9

May 12, 2026

NAS-Bench CIFAR-100 201 (test)

0.65Kendall's Tau

9

May 12, 2026

NAS-Bench-201 CIFAR-10 (test)

0.64Kendall's Tau

9

May 12, 2026

5.82MAE

9

Feb 26, 2026

9.33MAE

9

Feb 26, 2026

Average across Tasks

1.51MAE

9

Feb 26, 2026

3.15MAE

9

Feb 26, 2026

Page 1 of 2

PreviousNext

Popular datasets

ARC WMT Sentiment MMLU AmericasNLP

Follow for updates

Join SOTA2 Contact