Granite Switch 4.1

Granite Switch 4.1: Chat, Decisions and Speech in One Endpoint

Granite Switch 4.1's 8B and 30B models combine chat, general and specialized decision-making, and speech-to-text in one hosted endpoint, adding a Decision adapter, a Guardian adapter that outperforms several baselines on safety decisions, and the top OpenASR speech model. VRAM use matches the base model, and KV cache reuse enables fast switching.