Best AI models for images and PDFs

Models that accept images as input, so they can read screenshots, photos, charts and scanned or PDF documents. Ranked by the overall index from Epoch AI’s independent tests.

Updated September 30, 2026

Our picks

#ModelOverall indexiInputiOutputiContexti
1
166.6
$10$501.1M
2
165.0
$10$501M
3
163.6
$10$501M
4
Claude Opus 5Anthropic
162.7
$5$251M
5
162.5
$30$1801.1M
6
162.0
$2$101.1M
7
159.3
$2$121.1M
8
GPT-5.5OpenAI
159.3
$5$301.1M
9
159.1
$30$1801.1M
10
158.3
$5$251M

per 1M tokens · USD

Quick answers

Which AI model is best for reading images and PDFs?
GPT-6 Astra from OpenAI leads this ranking (Overall index: 166.6). It costs $10 per million input tokens and $50 per million output tokens.
Is there a cheaper alternative?
GPT-5.6 Sol from OpenAI (Overall index: 162.0) costs $2 / $10 per million tokens, about 5.0× less than GPT-6 Astra.

How we rank

Scores come from tests run independently by Epoch AI (CC BY 4.0), not from the companies’ own announcements. Prices come from OpenRouter’s public API. Both are refreshed automatically several times a day, so this ranking follows new releases and price changes.

Other rankings

CodingMathScienceCheapFreeOpen sourceLong contextRankings →