Claude Code used by 80% of developers despite GPT-5.5 producing fewer bugs

According to Cubic's State of AI Coding 2026 report analyzing thousands of daily commits, GPT-5.5 produces fewer bugs per lines changed than any other model in their dataset, yet Claude Code remains the most popular coding agent with 80% developer adoption. Opus models are the most-used model family, with Claude Fable adopted by 30% of developers within 48 hours of release. The report notes that Opus 4.8 shows a higher bug rate than Opus 4.7, for which the authors have no firm explanation.

Topics

Agentic AIClaude CodeChatGPT

Sources

Go deeper

This intelligence is sourced automatically from public sources across the web and synthesised by the Prefactor AI pipeline. Stories are reviewed before publication.