OpenAI Codex CLI vs Cursor Agent: Benchmark Comparison

Independent benchmark data · Real published scores only

📊 SWE-bench Verified

🏆 Higher Score

OpenAI Codex CLI

OpenAI

68.4

Trust Score V2

95% CI: 64.9 – 71.9

View full profile →

VS

Cursor Agent

Cursor

63.9

Trust Score V2

95% CI: 60.4 – 67.4

View full profile →

Score Comparison

OpenAI Codex CLI

Cursor Agent

Trust Score

68.4

63.9

Functional Acc.

69.1

51.7

Reliability

63.7

65.2

Policy Compliance

90.1

91.5

Key Metrics

Metric	OpenAI Codex CLI	Cursor Agent
Trust Score V2	68.4	63.9
Functional Accuracy	69.1	51.7
Reliability Score	63.7	65.2
Policy Compliance	90.1	91.5
SWE-bench Pass@1	0.7%	0.5%
Benchmark	SWE-bench Verified	SWE-bench Verified
Last Evaluated	Mar 13, 2026	Mar 13, 2026
Model Base	o3	Claude 3.5 Sonnet