Skip to content

Benchmark

IDP Core Bench

OCR benchmarks measure how well models read text in images and documents. Scores are percentages.

Models measured
24
Scores recorded
24
Sources
1
IQ weight
1.20

Score distribution

68.9%
Mean
74.5%
Median
17.9%
Std dev
54.9%
10th percentile
81.8%
90th percentile
0%
Min
0%89.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.20 for chat models, measured over 23 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard