Claude Opus 4.5 outscores human engineering candidates

Monday 24 November 2025 · significance ●●●○○

Anthropic says Opus 4.5 scored higher than any human candidate on its take-home engineering exam, with 80.9% on SWE-bench Verified.

What happened and why it matters

Opus 4.5 was the first model above 80% on SWE-bench Verified. Exceeding every human applicant on a performance-engineering exam used to hire Anthropic's own engineers is a direct signal for AI-driven AI development.

Sources

Introducing Claude Opus 4.5 Anthropic · 24 Nov 2025

Effect on the Gapnium Coefficient

+0.0054

0.2700 → 0.2754

Adjustments

gapnium.autonomy +0.015 New high on real-world software engineering.
gapnium.recursion +0.012 Out-performs human candidates on the lab's own engineering hiring exam.

Review

Last reviewed by AI agent: 7 Nov 2026
Model: seed:claude-opus-5-5
Agent confidence: 0.75
Human reviewed: —

Initial historical seed. Scores assigned by the methodology rubric; awaiting first scheduled re-review.

Feedback on this scoring

Hover the chart to read values · click a marker to open its articles