Granite 4.2 3B vs Llama 4 Scout

Which is better for the money: Granite 4.2 3B by IBM or Llama 4 Scout by Meta? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-26.

Clear winner

IBMGranite 4.2 3BOn the open-weights frontier

Intelligence: 9.10
Total cost: $14.9

MetaLlama 4 ScoutOff the frontier

Intelligence: 6.50
Total cost: $73.6

Verdict

Pick Granite 4.2 3B

Llama 4 Scout vs Granite 4.2 3B: 28.6% less intelligence, 393.4% more cost to run the AA benchmark, and 91.1% faster end-to-end.

Metric	Granite 4.2 3B	Llama 4 Scout	Difference
Intelligence (higher is better)	9.1	6.5	−28.6% Llama 4 Scout is 28.6% less intelligent
Total AA cost (lower is better)	$14.91	$73.56	+393.4% Llama 4 Scout is 4.9× more expensive
End-to-end time (lower is better)	55.67s	4.97s	−91.1% Llama 4 Scout is 91.1% faster
Price per 1M tokens (in / out)	$0.03 / $0.12	$0.18 / $0.66
Input modalities	Text	Text, Image

Open this matchup in the interactive picker Granite 4.2 3B: what beats it Llama 4 Scout: what beats it

Today's Pareto frontier

Against the frontier

Granite 4.2 3B is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
Llama 4 Scout is off the frontier: at the same or lower cost, GPT-6 Luna (Xhigh) scores 34.60 for $66.8.

Granite 4.2 3BLlama 4 ScoutToday's Pareto frontierOpen-weights frontierEvery other model

Matchup

Compare something else

Agents: Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/granite-4-2-3b/llama-4-scout · Markdown: /api/v1/compare/granite-4-2-3b/llama-4-scout?format=md · llms.txt · agent skill