GPT-OSS 20B vs Qwen2.5-VL-32B

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

Quick Verdict

Qwen2.5-VL-32B wins overall with a score of 34 vs 22 (12 point difference).Qwen2.5-VL-32B wins 4 out of 4 categories.

Knowledge

Qwen2.5-VL-32B

GPT-OSS 20B

28.8

Qwen2.5-VL-32B

40.8

31
MMLU
43
30
GPQA
42
28
SuperGPQA
40
26
OpenBookQA
38

Coding

Qwen2.5-VL-32B

GPT-OSS 20B

23

Qwen2.5-VL-32B

35

23
HumanEval
35

Mathematics

Qwen2.5-VL-32B

GPT-OSS 20B

30

Qwen2.5-VL-32B

42

31
AIME 2023
43
33
AIME 2024
45
32
AIME 2025
44
27
HMMT Feb 2023
39
29
HMMT Feb 2024
41
28
HMMT Feb 2025
40
30
BRUMO 2025
42

Reasoning

Qwen2.5-VL-32B

GPT-OSS 20B

28

Qwen2.5-VL-32B

40

29
SimpleQA
41
27
MuSR
39

Frequently Asked Questions

Which is better, GPT-OSS 20B or Qwen2.5-VL-32B?

Qwen2.5-VL-32B scores higher overall with 34 vs 22, a difference of 12 points across all benchmarks.

Which is better for knowledge tasks, GPT-OSS 20B or Qwen2.5-VL-32B?

Qwen2.5-VL-32B leads in knowledge tasks with an average score of 40.8 vs 28.8.

Which is better for coding, GPT-OSS 20B or Qwen2.5-VL-32B?

Qwen2.5-VL-32B leads in coding with an average score of 35 vs 23.

Which is better for math, GPT-OSS 20B or Qwen2.5-VL-32B?

Qwen2.5-VL-32B leads in math with an average score of 42 vs 30.

Which is better for reasoning, GPT-OSS 20B or Qwen2.5-VL-32B?

Qwen2.5-VL-32B leads in reasoning with an average score of 40 vs 28.