Back to Benchmarks
GPT-4.1 vs DeepSeek R1 Distill Llama 70B
GPT-4.1
OpenAI
vs
DeepSeek R1 Distill Llama 70B
DeepSeek
via DeepInfra
GPT-4.1 wins 7 of 11
Index Scores
26.3
+10.3
Intelligence
16.0
21.8
+10.4
Coding
11.4
34.7
Math
+19.0
53.7
Benchmarks
80.6
+1.1
MMLU Pro
79.5
66.6
+26.4
GPQA
40.2
4.6
HLE
+1.5
6.1
45.7
+19.1
LiveCode
26.6
38.1
+6.9
SciCode
31.2
91.3
MATH-500
+2.2
93.5
43.7
AIME
+23.3
67.0
Speed
44
+2
TPS
42
Shared
Winning delta
crafted by
bart stefanski