Back to Benchmarks
Qwen3.5 397B A17B (Reasoning) vs Grok 4
Qwen3.5 397B A17B (Reasoning)
Alibaba
vs
Grok 4
xAI
Grok 4 wins 6 of 11
Index Scores
45.0
+3.5
Intelligence
41.5
41.3
+0.8
Coding
40.5
0.0
Math
+92.7
92.7
Benchmarks
0.0
MMLU Pro
+86.6
86.6
89.3
+1.6
GPQA
87.7
27.3
+3.4
HLE
23.9
0.0
LiveCode
+81.9
81.9
42.0
SciCode
+3.7
45.7
0.0
MATH-500
+99.0
99.0
0.0
AIME
+94.3
94.3
Speed
53
+8
TPS
45
Shared
Winning delta
crafted by
bart stefanski