Not by version attribution: in 3 runs, Claude Opus 5 places its react-router boundary at 7.9.0 (2025-09-12) — below react-router 8.0.0 (2026-06-17), the first release of react-router 8. That is ~8 months below the training cutoff the subject stated in those runs (2026-05). 1 finding is currently charged against Claude Opus 5 on react-router (1x S2 silently-wrong), each reproduced in a published run and checked against react-router's own release notes. Read the boundary precisely: it is the newest release whose contents the model can correctly attribute to that release, not the newest react-router feature it can use. Past it a model often writes working code with a newer API while naming the wrong release for it.
Answer class below, decided by one rule applied to every page
of this kind: every measured boundary is below the first release of the major named in the question. Every figure below is read from
the dataset at build time; nothing on this page is written by hand.
3 measurements of this pair, all giving the same boundary. No two of them were drawn from an identical prompt, so the spread mixes prompt differences with draw variance.
react-router published 9 minor or major releases in the twelve months before this model’s stated cutoff, which is the scale a spread should be read against.
“Chargeable” means the change was published before this model’s own stated cutoff, so it had the opportunity to know it.
The react-router correction pack states what is true now for each corrected fact, with a primary-source citation, as markdown you can paste into a rules file (raw). What a correction pack measurably changed when one was tested — a pre-registered run on prisma and zod — is on the benchmark page, including where it changed nothing.
All of them at once: Which Claude model knows react-router 8 best?
Other subjects on react-router:
Claude Opus 5 on the other libraries: