Codesota · Models · ViT-B/16Google2 results · 2 benchmarks
Model card

ViT-B/16.

Googleopen-source86M paramsVision TransformerApache 2.0

Base variant. 81.2% with ImageNet-21k pretraining.

§ 02 · Benchmarks

Every benchmark ViT-B/16 has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01CIFAR-100Computer Vision · Image Classificationaccuracy91.5%#7/10source ↗
02ImageNet-1KComputer Vision · Image Classificationtop-1-accuracy81.2%#16/20source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area

Where ViT-B/16 actually performs.

Computer Vision
2
benchmarks
avg rank #11.5
§ 05 · Related models

Other Google models scored on Codesota.

Gemini 1.5 Pro
21 results
Gemini 2.5 Pro
16 results
Gemini 3 Pro
Undisclosed params · 13 results
Gemini-3.1-Pro
11 results
Gemma-3-27b
27B params · 11 results
Gemma 3 (27B, IT)
9 results
Gemma-2-27b-it
9 results
gemma-3-12b-it
9 results
§ 06 · Sources & freshness

Where these numbers come from.

huggingface
1
result
google-research
1
result
0 of 2 rows marked verified.