Skip to main content
CodingAgentBench

Codex CLI

openai's open-source agentic CLI for BYO-model autonomous one-shot tasks.

v0.140.0 upstream ↗

Watch it run — v0.140.0

Version history

Version Composite Pass-rate Cells
v0.140.0 0.550 18.6% 350 rows →

Performance is tracked per pinned upstream version. New releases are auto-rebenched and a fresh "watch it run" is captured — the trend grows here.