Back to Benchmarks

Llama 3.3 Instruct 70B vs GPT-4.1 nano

Llama 3.3 Instruct 70B

Metavia Groq

vs

GPT-4.1 nano

OpenAI

Llama 3.3 Instruct 70B(Groq) wins 6 of 11

Index Scores

14.5
+1.5
Intelligence
13.0
10.7
Coding
+0.5
11.2
7.7
Math
+16.3
24.0

Benchmarks

71.3
+5.6
MMLU Pro
65.7
49.8
GPQA
+1.4
51.2
4.0
+0.1
HLE
3.9
28.8
LiveCode
+3.8
32.6
26.0
+0.1
SciCode
25.9
77.3
MATH-500
+7.5
84.8
30.0
+6.3
AIME
23.7

Speed

144
+109
TPS
35
Shared
Winning delta

crafted by bart stefanski