Foundation Model
A large model trained on broad data that is adapted (via prompting, fine-tuning, or RAG) for downstream tasks.
What is Foundation Model?
Foundation models are the generic substrate of modern AI. Enterprises rarely use them raw; they layer retrieval, routing, and governance on top. For high-volume internal tasks, a smaller specialised model often outperforms a generic foundation model on cost, latency, and energy. See Small Language Models in Enterprise AI.
What is an example of Foundation Model?
A broadly trained language model becomes the base for a legal research assistant when the application adds a private case-law index, citation checks, access controls, and a workflow tailored to legal questions.
How is Foundation Model different from related concepts?
A foundation model is the reusable model layer. A general-purpose AI system is a complete system that may use such a model, while a domain application is configured for a particular business purpose.
What should enterprises evaluate for Foundation Model?
- Benchmark the model on representative tasks, languages, safety cases, and long-context workloads instead of generic leaderboards.
- Review license, provenance, deployment rights, support horizon, security posture, and documented limitations.
- Evaluate the complete application after retrieval, prompts, tools, and guardrails are added.
Related terms
Putting Foundation Model to work?
VDF AI runs governed AI agents on your own infrastructure — on-premises, sovereign cloud, or air-gapped. Book a working session to map the architecture.
Talk to VDF AI