Tools
I tried a 15MB, 30 page text-heavy PDF and Opus 4.7 reported 60,934 tokens while 4.6 reported 56,482 - that's a 1.08x multiplier, significan…
Claude Opus 4.7 uses approximately 8% more tokens than Opus 4.6 when processing the same text-heavy PDF document, as demonstrated by a test comparing tokenization of a 15MB, 30-page PDF (60,934 tokens
Claude Opus 4.7 uses approximately 8% more tokens than Opus 4.6 when processing the same text-heavy PDF document, as demonstrated by a test comparing tokenization of a 15MB, 30-page PDF (60,934 tokens vs 56,482 tokens respectively). This difference in token counting between model versions may impact API costs and context window usage for users working with PDF documents.
Related
- Important to note: that 3x increase for images is entirely due to Opus 4.7 being able to handle higher resolutions. I tried that again with …
- Opus 4.7 uses more thinking tokens, so we've increased rate limits for all subscribers to make up for it. Enjoy!
Source: Simon Willison (X) | 2026-04-20