Back to Benchmarks

GPT-4o (Aug '24) vs DeepSeek R1 Distill Llama 70B

GPT-4o (Aug '24)

OpenAI

vs

DeepSeek R1 Distill Llama 70B

DeepSeekvia DeepInfra

DeepSeek R1 Distill Llama 70B(DeepInfra) wins 6 of 11

Index Scores

18.6
+2.6
Intelligence
16.0
16.6
+5.2
Coding
11.4
0.0
Math
+53.7
53.7

Benchmarks

0.0
MMLU Pro
+79.5
79.5
52.1
+11.9
GPQA
40.2
2.9
HLE
+3.2
6.1
31.7
+5.1
LiveCode
26.6
33.1
+1.9
SciCode
31.2
79.5
MATH-500
+14.0
93.5
11.7
AIME
+55.3
67.0

Speed

14
TPS
+28
42
Shared
Winning delta

crafted by bart stefanski