Fetching from the wire…
Research2026-04-21 · source-backed
Researchers analyzed 61,837 GitHub Actions runs from 2,355 repos triggered by PRs from Claude, Devin, Cursor, Copilot, and Codex. Substantial differences in pass rates across bots. This is the first empirical data on how AI-generated code actually performs under real CI/CD constraints.
Each link below shares sources, entities, or timing with this story.
Cursor benchmarked against Codex / Shared entities / Same source domain / Shared topic / What happened next
Linked by a graph relationship (Cursor benchmarked against Codex); both cover CLAUDE, Copilot, Cursor, Devin; reported by the same outlet (arxiv.org).
Linked by a graph relationship (Cursor benchmarked against Codex); both cover CLAUDE, Copilot, Cursor, Devin; reported by the same outlet (arxiv.org).
Codex competes with Claude Code / Shared entities / Shared topic / What happened next
Linked by a graph relationship (Codex competes with Claude Code); both cover CLAUDE, Codex, Copilot, Cursor; overlapping topics (claude, code, codex, copilot, cursor).
Codex competes with Claude Code / Shared entities / Shared topic / Earlier coverage
Linked by a graph relationship (Codex competes with Claude Code); both cover Codex, Cursor, PRs; overlapping topics (actually, claude, code, coding, cursor).
Linked by a graph relationship (Codex competes with Claude Code); both cover Codex, Copilot, Cursor; overlapping topics (claude, code, codex, coding, copilot).
Codex competes with Gemini / Shared entities / Same source domain / Shared topic / What happened next
Linked by a graph relationship (Codex competes with Gemini); both cover CLAUDE, Codex, Copilot; reported by the same outlet (arxiv.org).
Codex competes with Claude Code / Shared entities / Shared topic / What happened next
Linked by a graph relationship (Codex competes with Claude Code); both cover Codex, Copilot, Cursor; overlapping topics (claude, code, codex, coding, copilot).
Linked by a graph relationship (Codex competes with Claude Code); both cover CLAUDE, Codex, Cursor; overlapping topics (claude, code, codex, coding, cursor).