benchlm.ai

GPT-OSS 20B vs o4-mini (high)

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

Quick Verdict

o4-mini (high) wins overall with a score of 65 vs 22 (43 point difference).o4-mini (high) wins 4 out of 4 categories.

GPT-OSS 20B

CreatorOpenAI
TypeOpen Weight
Context128K
Score22

o4-mini (high)

CreatorOpenAI
TypeProprietary
Context200K
Score65

Knowledge

o4-mini (high)

GPT-OSS 20B

28.8

o4-mini (high)

80.5

31
MMLU
82
30
GPQA
82
28
SuperGPQA
80
26
OpenBookQA
78

Coding

o4-mini (high)

GPT-OSS 20B

23

o4-mini (high)

74

23
HumanEval
74

Mathematics

o4-mini (high)

GPT-OSS 20B

30

o4-mini (high)

82

31
AIME 2023
83
33
AIME 2024
85
32
AIME 2025
84
27
HMMT Feb 2023
79
29
HMMT Feb 2024
81
28
HMMT Feb 2025
80
30
BRUMO 2025
82

Reasoning

o4-mini (high)

GPT-OSS 20B

28

o4-mini (high)

79

29
SimpleQA
80
27
MuSR
78

Frequently Asked Questions

Which is better, GPT-OSS 20B or o4-mini (high)?

o4-mini (high) scores higher overall with 65 vs 22, a difference of 43 points across all benchmarks.

Which is better for knowledge tasks, GPT-OSS 20B or o4-mini (high)?

o4-mini (high) leads in knowledge tasks with an average score of 80.5 vs 28.8.

Which is better for coding, GPT-OSS 20B or o4-mini (high)?

o4-mini (high) leads in coding with an average score of 74 vs 23.

Which is better for math, GPT-OSS 20B or o4-mini (high)?

o4-mini (high) leads in math with an average score of 82 vs 30.

Which is better for reasoning, GPT-OSS 20B or o4-mini (high)?

o4-mini (high) leads in reasoning with an average score of 79 vs 28.

Related Comparisons

GPT-OSS 20B vs GPT-5.4GPT-OSS 20B vs Gemini 3.1 ProGPT-OSS 20B vs Claude Opus 4.6GPT-OSS 20B vs GPT-5.3 CodexGPT-OSS 20B vs Grok 4.1GPT-OSS 20B vs GPT-5.2