IBM opens Granite 4.2 under Apache 2.0: three sizes, half a million tokens of context
The models were trained from scratch on 15 trillion tokens and can toggle thinking on and off. The speech model transcribes three hours of audio in one second.
Models