The practical experience is matching the benchmark data Anthropic published.
If Opus 4.7 is this good, I'm excited to see what Mythos brings.
Benchmarks data from Anthropic below.
First published on LinkedIn.
I took a step back last week to consider where we're investing effort into AI as a business, broadly categorising them into two areas: efficiency and innovation.
I was at Hampden's market briefing at Fishmongers’ Hall this morning, and one comment has really stuck with me.
By the end of it, I’m feeling super recharged. The difference was leaning into the things I felt compelled to do, especially helping someone else.
What happened this week, what I'm thinking about, and one takeaway for yours.
GEThttps://sam-page.com/thinking/weve-been-using-claude-opus-4-7-since-it-launched-last-week.mdtext/markdown; charset=utf-8
--- title: "We've been using Claude Opus 4.7 since it launched last week." url: https://sam-page.com/thinking/weve-been-using-claude-opus-4-7-since-it-launched-last-week date: 2026-04-21 author: Sam Page type: Post or article categories: [Emerging tech and AI] reading_time: 1 min summary: "The practical experience is matching the benchmark data Anthropic published." first_published: https://www.linkedin.com/feed/update/urn%3Ali%3Ashare%3A7452233558254837760 --- # We've been using Claude Opus 4.7 since it launched last week. The practical experience is matching the benchmark data Anthropic published.  - Coding is a genuine leap forward compared to 4.6 and other models. - Multi-step reasoning is working noticeably better. - Visual capabilities are a step up. If Opus 4.7 is this good, I'm excited to see what Mythos brings. Benchmarks data from Anthropic below.
