3B, 8B, 30B
128K tokens
Apache 2.0
Official open-weight Granite 4.1 family release spanning 3B, 8B, and 30B text models, each published in both base and chat-capable variants. Public Hugging Face configs for the family indicate dense decoder-only GraniteForCausalLM architecture with 131,072 maximum position embeddings, Granite chat templating for conversational and tool-oriented prompting, and Apache 2.0 licensing across the release. IBM positions Granite as an enterprise-oriented model family emphasizing practical instruction following, deployment flexibility, and compatibility with standard open-source inference stacks. The 4.1 release extends that line with multiple scale points so teams can trade off quality, latency, and serving cost while staying within one architecture family and prompt format.