Claude Opus 4.7 costs 20–30% more per session

I Measured Claude 4.7's New Tokenizer. Here's What It Costs You.

claudecodecamp.com

April 17, 2026

1 min read

🔥🔥🔥🔥🔥

71/100

Summary

Claude 4.7's new tokenizer uses 1.47 times more tokens than previous versions, exceeding the documentation estimate of 1.0–1.35x. This increase impacts the cost of processing content.

Read original article

View story →

Coverage & discussions

Hacker News · claudecodecamp.com

Claude Opus 4.7 costs 20–30% more per session ↗

Community discussions

Hacker News discussion ↗

Historical count: 5 · 5 saved in sample · Historical sample

Analysis ready

Community Sentiment

Mixed

Positives

For specific high-stakes tasks, even a small increase in model intelligence can justify significantly higher costs, highlighting the importance of performance in critical applications.
Some users report that Opus 4.7 has improved speed and efficiency, leading to lower session usage despite the higher costs, which could enhance productivity.

Concerns

The 20-30% cost increase for Opus 4.7 raises concerns about diminishing returns in model quality, as many users may not perceive a meaningful difference in their daily tasks.
There is skepticism about whether Opus 4.7 truly offers a substantial upgrade over 4.6, with some users suggesting it may not justify the higher expense.
The rapid price increase may indicate that Anthropic is struggling with high operational costs, which could impact the sustainability of their model offerings.