Definition: A foundation model is a large AI model trained on broad data at scale that can be adapted to many downstream tasks, such as chat, coding, search or image analysis, through prompting or fine-tuning.
The term was popularized by Stanford researchers in 2021. Examples include the GPT, Claude, Gemini and Llama families, along with models for images, audio and code. Large language models are the best-known type.
Companies rarely train foundation models. They access them through APIs and cloud platforms, or self-host open-weight models, then adapt them with prompts, RAG, tool access or fine-tuning.
Selection usually weighs quality on your own evals, cost, latency, context length, data residency, deployment options and vendor terms. Many teams use more than one model for different tasks, and keep their application code model-agnostic so they can switch as newer models are released and prices change.
Ryz AI pod teams build systems on Anthropic, OpenAI and other models in your cloud and repositories, working alongside your team.
Thanks — your message has been sent. We’ll get back to you soon.
Something went wrong while sending your message. Please try again or email info@ryzlabs.com.