benchlm.ai

Claude 3.5 Sonnet vs GPT-5.4

Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.

Quick Verdict

GPT-5.4 wins overall with a score of 88 vs 55 (33 point difference).GPT-5.4 wins 4 out of 4 categories.

Claude 3.5 Sonnet

CreatorAnthropic
TypeProprietary
Context200K
Score55

GPT-5.4

CreatorOpenAI
TypeProprietary
Context1M
Score88

Knowledge

GPT-5.4

Claude 3.5 Sonnet

63.5

GPT-5.4

96

65
MMLU
99
65
GPQA
97
63
SuperGPQA
95
61
OpenBookQA
93

Coding

GPT-5.4

Claude 3.5 Sonnet

57

GPT-5.4

91

57
HumanEval
91

Mathematics

GPT-5.4

Claude 3.5 Sonnet

64

GPT-5.4

97.1

65
AIME 2023
99
67
AIME 2024
99
66
AIME 2025
98
61
HMMT Feb 2023
95
63
HMMT Feb 2024
97
62
HMMT Feb 2025
96
64
BRUMO 2025
96

Reasoning

GPT-5.4

Claude 3.5 Sonnet

62

GPT-5.4

94

63
SimpleQA
95
61
MuSR
93

Frequently Asked Questions

Which is better, Claude 3.5 Sonnet or GPT-5.4?

GPT-5.4 scores higher overall with 88 vs 55, a difference of 33 points across all benchmarks.

Which is better for knowledge tasks, Claude 3.5 Sonnet or GPT-5.4?

GPT-5.4 leads in knowledge tasks with an average score of 96 vs 63.5.

Which is better for coding, Claude 3.5 Sonnet or GPT-5.4?

GPT-5.4 leads in coding with an average score of 91 vs 57.

Which is better for math, Claude 3.5 Sonnet or GPT-5.4?

GPT-5.4 leads in math with an average score of 97.1 vs 64.

Which is better for reasoning, Claude 3.5 Sonnet or GPT-5.4?

GPT-5.4 leads in reasoning with an average score of 94 vs 62.

Related Comparisons

Claude 3.5 Sonnet vs Gemini 3.1 ProClaude 3.5 Sonnet vs Claude Opus 4.6Claude 3.5 Sonnet vs GPT-5.3 CodexClaude 3.5 Sonnet vs Grok 4.1Claude 3.5 Sonnet vs GPT-5.2Claude 3.5 Sonnet vs GPT-5.2-Codex