A foundation model is a large AI model trained on broad data that can be adapted to many downstream tasks through prompting or fine-tuning. Large language models are the best-known example. IBM's Granite family is a set of foundation models built for enterprise use with attention to licensing, transparency, and governance.
Also known as: large language model, LLM
The right model balances accuracy, latency, cost, and governance for a given workflow. Enterprises increasingly mix models — open and proprietary — rather than betting on one.
On IBM watsonx.ai we select and tune the model that fits each use case, including Granite where its enterprise and governance profile is the best fit.
IBM's enterprise studio to build, train, tune, and deploy AI models — including the Granite foundation models — under enterprise controls.
Read definition →ArchitectureA technique that grounds an AI model's responses in trusted external data retrieved at query time — improving accuracy and reducing hallucination.
Read definition →Agentic AIAI systems that take action — applying logic, making decisions, and running multi-step workflows end to end across enterprise systems, with human oversight.
Read definition →