Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
GPT-5.3 Codex wins overall with a score of 85 vs 46 (39 point difference).GPT-5.3 Codex wins 4 out of 4 categories.
Claude 3 Haiku
54.5
GPT-5.3 Codex
96
Claude 3 Haiku
48
GPT-5.3 Codex
95
Claude 3 Haiku
55
GPT-5.3 Codex
97.1
Claude 3 Haiku
53
GPT-5.3 Codex
94
GPT-5.3 Codex scores higher overall with 85 vs 46, a difference of 39 points across all benchmarks.
GPT-5.3 Codex leads in knowledge tasks with an average score of 96 vs 54.5.
GPT-5.3 Codex leads in coding with an average score of 95 vs 48.
GPT-5.3 Codex leads in math with an average score of 97.1 vs 55.
GPT-5.3 Codex leads in reasoning with an average score of 94 vs 53.