Granite 4.2 30B

IBM llm

IBM dense reasoning model fine-tuned from Granite-4.1-30B-Base with thinking, non-thinking and low-effort modes.

overall ⌄

36.9

92 of 125 ranked

price · $/M tokens

self-hosted

open weights, no first-party API

to run it yourself ⌄

18 GB

minimum · 72 GB recommended

Benchmarks

What its maker published, and what anyone else measured. Every figure links the document it came from.

benchmarkpublishedmeasured
MMLU-Pro77.6 thinking—
AIME 202589.17 thinking—
SWE-bench57 —
SWE-bench Pro33.29 —

Lineage

view in full graph →

Metadata

orgIBM
released2026-08-25
params30B
licenseapache-2.0
hf idibm-granite/granite-4.2-30b