Claude Code vs OpenAI Codex CLI: Benchmark Comparison

Independent benchmark data · Real published scores only

📊 SWE-bench Verified

🏆 Higher Score

Claude Code

Anthropic

71.5

Trust Score V2

95% CI: 68.0 – 75.0

View full profile →

VS

OpenAI Codex CLI

OpenAI

68.4

Trust Score V2

95% CI: 64.9 – 71.9

View full profile →

Score Comparison

Claude Code

OpenAI Codex CLI

Trust Score

71.5

68.4

Functional Acc.

72.5

69.1

Reliability

70.1

63.7

Policy Compliance

94.2

90.1

Key Metrics

Metric	Claude Code	OpenAI Codex CLI
Trust Score V2	71.5	68.4
Functional Accuracy	72.5	69.1
Reliability Score	70.1	63.7
Policy Compliance	94.2	90.1
SWE-bench Pass@1	0.7%	0.7%
Benchmark	SWE-bench Verified	SWE-bench Verified
Last Evaluated	Mar 13, 2026	Mar 13, 2026
Model Base	Claude Opus 4	o3