OpenAI Codex CLI vs Google Antigravity: Benchmark Comparison

Independent benchmark data · Real published scores only
📊 SWE-bench Verified
OpenAI Codex CLI
OpenAI
68.4
Trust Score V2
95% CI: 64.9 – 71.9
View full profile →
VS
🏆 Higher Score
Google Antigravity
Google
72.4
Trust Score V2
95% CI: 69.0 – 75.9
View full profile →
Score Comparison
OpenAI Codex CLI
Google Antigravity
Trust Score
68.4
72.4
Functional Acc.
69.1
76.2
Reliability
63.7
71.5
Policy Compliance
90.1
91.4
Key Metrics
Metric OpenAI Codex CLI Google Antigravity
Trust Score V2 68.4 72.4
Functional Accuracy 69.1 76.2
Reliability Score 63.7 71.5
Policy Compliance 90.1 91.4
SWE-bench Pass@1 0.7% 0.8%
Benchmark SWE-bench Verified SWE-bench Verified
Last Evaluated Mar 13, 2026 Mar 17, 2026
Model Base o3 Gemini 3 Pro

Need a procurement-grade evaluation report?

Get cost-of-failure modeling, compliance validation, and a certified comparison report for OpenAI Codex CLI and Google Antigravity — built for enterprise procurement decisions.

Request Evaluation Report →