Neura Intelligence Index

AI Model Benchmarks

Back to Leaderboard

IBM Granite 4.2 8B vs Llama 4 Scout

I

IBM Granite 4.2 8B

IBM 🇺🇸

37.1

#250—

M

Meta 🇺🇸

37.4

#247—

Category Comparison

Benchmark-by-Benchmark

reasoning

Benchmark	IBM Granite 4.2 8B	Llama 4 Scout
GPQA Diamond	64.1	57.2

math

Benchmark	IBM Granite 4.2 8B	Llama 4 Scout
AIME 2025	86.7	—

coding

Benchmark	IBM Granite 4.2 8B	Llama 4 Scout
SWE-bench Verified	47.7	—
SWE-Bench Pro	19.1	—
SciCode	36.1	—

multimodal

Benchmark	IBM Granite 4.2 8B	Llama 4 Scout
MMMU	—	69.4

Category Winners

coding

I

IBM Granite 4.2 8B

11.8

math

I

IBM Granite 4.2 8B

64.1

reasoning

I

IBM Granite 4.2 8B

42.1

multimodal

M

Llama 4 Scout

53.7

IBM Granite 4.2 8B vs Llama 4 Scout — Benchmark Comparison | Neura Market