Claude 3 Haiku vs GPT-OSS 120B

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

Quick Verdict

Claude 3 Haiku wins overall with a score of 46 vs 42 (4 point difference).Claude 3 Haiku wins 4 out of 4 categories.

Knowledge

Claude 3 Haiku

Claude 3 Haiku

54.5

GPT-OSS 120B

48.8

56
MMLU
51
56
GPQA
50
54
SuperGPQA
48
52
OpenBookQA
46

Coding

Claude 3 Haiku

Claude 3 Haiku

48

GPT-OSS 120B

43

48
HumanEval
43

Mathematics

Claude 3 Haiku

Claude 3 Haiku

55

GPT-OSS 120B

50

56
AIME 2023
51
58
AIME 2024
53
57
AIME 2025
52
52
HMMT Feb 2023
47
54
HMMT Feb 2024
49
53
HMMT Feb 2025
48
55
BRUMO 2025
50

Reasoning

Claude 3 Haiku

Claude 3 Haiku

53

GPT-OSS 120B

48

54
SimpleQA
49
52
MuSR
47

Frequently Asked Questions

Which is better, Claude 3 Haiku or GPT-OSS 120B?

Claude 3 Haiku scores higher overall with 46 vs 42, a difference of 4 points across all benchmarks.

Which is better for knowledge tasks, Claude 3 Haiku or GPT-OSS 120B?

Claude 3 Haiku leads in knowledge tasks with an average score of 54.5 vs 48.8.

Which is better for coding, Claude 3 Haiku or GPT-OSS 120B?

Claude 3 Haiku leads in coding with an average score of 48 vs 43.

Which is better for math, Claude 3 Haiku or GPT-OSS 120B?

Claude 3 Haiku leads in math with an average score of 55 vs 50.

Which is better for reasoning, Claude 3 Haiku or GPT-OSS 120B?

Claude 3 Haiku leads in reasoning with an average score of 53 vs 48.