Back to Benchmarks
GPT-4.1 vs DeepSeek R1 Distill Qwen 32B
GPT-4.1
OpenAI
vs
DeepSeek R1 Distill Qwen 32B
DeepSeek
via NextBit
GPT-4.1 wins 7 of 11
Index Scores
26.3
+9.1
Intelligence
17.2
21.8
+21.8
Coding
0.0
34.7
Math
+28.3
63.0
Benchmarks
80.6
+6.7
MMLU Pro
73.9
66.6
+5.1
GPQA
61.5
4.6
HLE
+0.9
5.5
45.7
+18.7
LiveCode
27.0
38.1
+0.5
SciCode
37.6
91.3
MATH-500
+2.8
94.1
43.7
AIME
+25.0
68.7
Speed
44
+20
TPS
24
Shared
Winning delta
crafted by
bart stefanski