Back to Benchmarks
gpt-oss-120B (high) vs Llama 3.3 Instruct 70B
gpt-oss-120B (high)
OpenAI
via Groq
vs
Llama 3.3 Instruct 70B
Meta
via Groq
gpt-oss-120B (high)
(Groq)
wins 9 of 11
Index Scores
33.3
+18.8
Intelligence
14.5
28.6
+17.9
Coding
10.7
93.4
+85.7
Math
7.7
Benchmarks
80.8
+9.5
MMLU Pro
71.3
78.2
+28.4
GPQA
49.8
18.5
+14.5
HLE
4.0
87.8
+59.0
LiveCode
28.8
38.9
+12.9
SciCode
26.0
0.0
MATH-500
+77.3
77.3
0.0
AIME
+30.0
30.0
Speed
349
+205
TPS
144
Shared
Winning delta
crafted by
bart stefanski