SWE-Explore benchmark released, testing 848 bug-fixing tasks across 203 open-source projects

Codex 5.3 reported 95% reliability by same user

Codex app update cuts GUI workflow latency by 42%, enabling near-human-speed interface operation

Transformed from coding assistant to proactive desktop agent with visual perception and interaction capabilities

Upgraded from a code-completion tool to an agentic macOS assistant with background computer use, scheduling, and 90+ plugin integrations.

Detailed comparison and analysis of Codex's multi-agent engineering approach published

Released as native Windows application, shifting from cloud-based GitHub Copilot service