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

Handwritten text recognition research benchmarks – Page 2 · SOTA2 Research

Handwritten text recognition

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

HANDS-VNOnDB (test)

0.035CER

5

Feb 26, 2026

RIMES line level (test)

2.19CER

5

Feb 26, 2026

RIMES (k-means split (Target))

29.9CER

4

Mar 5, 2026

RIMES Source (k-means split)

3.6CER

4

Mar 5, 2026

RIMES Lexicon split (Target)

12.9CER

4

Mar 5, 2026

RIMES Lexicon (Source)

3.3CER

4

Mar 5, 2026

IAM (k-means split (Source))

6.1CER

4

Mar 5, 2026

IAM Lexicon split (Source)

6CER

4

Mar 5, 2026

6.59CER (%)

4

Feb 26, 2026

7.93CER

4

Feb 26, 2026

0.48CER

4

Feb 26, 2026

0.013CER

4

Feb 26, 2026

2.02CER

4

Feb 26, 2026

1.3CER

4

Feb 26, 2026

Esposalles (test)

Ours (VQA-augmented)

3.8WER

3

Mar 5, 2026

4.16CER

3

Feb 26, 2026

0.045CER

3

Feb 26, 2026

2.97CER

2

Feb 26, 2026

Muharaf Arabic (test)

11.7CER

2

Feb 26, 2026

Saint Gall (test)

7.25CER

1

Feb 26, 2026

4.5CER

1

Feb 26, 2026

BenthamR0 (test)

7.1CER

1

Feb 26, 2026

Digital Peter (test)

10.5CER

1

Feb 26, 2026

7.8CER

1

Feb 26, 2026

Ajami Fulfulde Hausa (test)

Pre-trained models from OPEN ITI

6,484CER

1

Feb 26, 2026

Page 2 of 3

Popular datasets

IAM RIMES READ Bentham Esposalles

Follow for updates

Join SOTA2 Contact