benchlm.ai

GLM-5 (Reasoning) vs GPT-OSS 20B

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

Quick Verdict

GLM-5 (Reasoning) wins overall with a score of 75 vs 22 (53 point difference).GLM-5 (Reasoning) wins 4 out of 4 categories.

GLM-5 (Reasoning)

CreatorZhipu AI
TypeOpen Weight
Context200K
Score75

GPT-OSS 20B

CreatorOpenAI
TypeOpen Weight
Context128K
Score22

Knowledge

GLM-5 (Reasoning)

GLM-5 (Reasoning)

93

GPT-OSS 20B

28.8

96
MMLU
31
94
GPQA
30
92
SuperGPQA
28
90
OpenBookQA
26

Coding

GLM-5 (Reasoning)

GLM-5 (Reasoning)

88

GPT-OSS 20B

23

88
HumanEval
23

Mathematics

GLM-5 (Reasoning)

GLM-5 (Reasoning)

96.6

GPT-OSS 20B

30

98
AIME 2023
31
99
AIME 2024
33
98
AIME 2025
32
94
HMMT Feb 2023
27
96
HMMT Feb 2024
29
95
HMMT Feb 2025
28
96
BRUMO 2025
30

Reasoning

GLM-5 (Reasoning)

GLM-5 (Reasoning)

91

GPT-OSS 20B

28

92
SimpleQA
29
90
MuSR
27

Frequently Asked Questions

Which is better, GLM-5 (Reasoning) or GPT-OSS 20B?

GLM-5 (Reasoning) scores higher overall with 75 vs 22, a difference of 53 points across all benchmarks.

Which is better for knowledge tasks, GLM-5 (Reasoning) or GPT-OSS 20B?

GLM-5 (Reasoning) leads in knowledge tasks with an average score of 93 vs 28.8.

Which is better for coding, GLM-5 (Reasoning) or GPT-OSS 20B?

GLM-5 (Reasoning) leads in coding with an average score of 88 vs 23.

Which is better for math, GLM-5 (Reasoning) or GPT-OSS 20B?

GLM-5 (Reasoning) leads in math with an average score of 96.6 vs 30.

Which is better for reasoning, GLM-5 (Reasoning) or GPT-OSS 20B?

GLM-5 (Reasoning) leads in reasoning with an average score of 91 vs 28.

Related Comparisons

GLM-5 (Reasoning) vs GPT-5.4GLM-5 (Reasoning) vs Gemini 3.1 ProGLM-5 (Reasoning) vs Claude Opus 4.6GLM-5 (Reasoning) vs GPT-5.3 CodexGLM-5 (Reasoning) vs Grok 4.1GLM-5 (Reasoning) vs GPT-5.2