Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Text Recognition on 30-page PG (test)
Loading...
1.05
CER (%)
PG fine-tuned model
0.6292
3.4696
6.31
9.1504
Mar 10, 2026
CER (%)
WER (%)
Updated 1mo ago
Evaluation Results
Method
Method
Links
CER (%)
WER (%)
PG fine-tuned model
2026.03
1.05
4.69
Transkribus (19th c. Greek)
2026.03
6.14
14.82
Pre-trained CRNN (with artificial noise)
2026.03
8.12
11.23
Tesseract (Greek)
2026.03
11.57
39.65
Feedback
Search any
task
Search any
task