Neura Intelligence Index

AI Model Benchmarks

Back to Leaderboard

Grok 4.5 vs Grok-4 Heavy

X

xAI 🇺🇸

86.7

#10—

X

xAI 🇺🇸

85.6

#13—

Category Comparison

Benchmark-by-Benchmark

reasoning

Benchmark	Grok 4.5	Grok-4 Heavy
GPQA Diamond	93	88.4
Humanity's Last Exam	—	50.7

coding

Benchmark	Grok 4.5	Grok-4 Heavy
SWE-Bench Pro	64.7	—

math

Benchmark	Grok 4.5	Grok-4 Heavy
AIME 2025	—	100

Category Winners

coding

X

Grok 4.5

82.8

math

X

Grok-4 Heavy

84.0

reasoning

X

Grok 4.5

90.6

Grok 4.5 vs Grok-4 Heavy — Benchmark Comparison | Neura Market