Hacker News

Granite 4.1: IBM’s 8B Model Matching 32B MoE

Granite-4.1-IBMs-8B-Model-Is-Competing-With-Models-Four-Times-Its-Size

HomeTechGranite 4.1: IBM’s 8B Model Is Competing With Models Four Times Its… Granite 4.1: IBM’s 8B Model Is Competing With Models Four Times Its Size By Mohit Geryani April 30, 2026 0 Last updated: April 30, 2026 Share FacebookTwitterPinterestWhatsApp – Advertisement – IBM just released Granite 4.1, a family of open source language models built specifically for enterprise use. Three sizes, Apache 2.0 licensed, trained on 15 trillion tokens with a level of pipeline obsession that’s worth understanding. One result in the benchmarks doesn’t make sense until you understand how they built it. The 8B model. Dense architecture, no MoE tricks, no extended reasoning chains. It matches or beats Granite 4.0-H-Small across basically every benchmark they ran. That older…

Continue reading at Hacker News firethering.com
Advertisement