Google

Creator of the Gemini and Gemma model families. Google's vision models combine strong multilingual understanding with fast inference via AI Studio and Vertex AI.

Models on OpenOCR

Available now

~3,710 pages / $1~$0.00027 per standard page

Lowest reference cost

Available Models

Gemini 3.1 Flash LiteLLM Vision

~821 pages / $1~$0.00122 per standard page

1156ms

Nano Banana 2.1LLM Vision

Nano Banana 2.1 appears to be a fictional or placeholder Google Gemini-based OCR engine slug that does not correspond to a known or publicly documented model. Based on the 'Nano' designation and Gemini branding, it may be intended to represent a lightweight, on-device variant of Google's Gemini vision capabilities with broad OCR support.

~175 pages / $1~$0.0057 per standard page

Gemini 4 ArgonLLM Vision

Gemini 4 Argon is Google's multimodal OCR engine leveraging the Gemini 4 model family to extract and interpret text across a wide range of document types, handwriting, and natural scene images. It stands out for its deep language understanding, layout-awareness, and broad multilingual support powered by Google's latest large-scale vision-language architecture.

~132 pages / $1~$0.0076 per standard page

Tesseract 5Open Source

Tesseract 5 with LSTM neural network — fast, reliable OCR for printed text.

500 pages / $1$0.002/page

420ms

Gemini 2.5 Flash-LiteLLM Vision

Gemini 2.5 Flash-Lite — OCR-capable vision model.

~3,710 pages / $1~$0.00027 per standard page

1652ms

Gemini 3.1 Flash Lite PreviewLLM Vision

Gemini 3.1 Flash Lite Preview — OCR-capable vision model.

~845 pages / $1~$0.00118 per standard page