Large Language Model (LLM)
A neural network trained on large text corpora to predict the next token given context, used for text generation, summarisation, classification, and reasoning tasks across enterprise software.
How it works
Large language models are the foundation of modern enterprise AI. They are trained on tens of billions of tokens of text and code, with parameter counts ranging from billions (Llama 3 8B, Qwen 7B) to hundreds of billions (GPT-4-class, Claude Opus). At inference, they take a sequence of input tokens (the prompt) and produce an output sequence one token at a time. The practical decision for UK businesses is rarely which LLM is most capable in isolation. It is which model can be deployed inside the firm under the right data-handling architecture: a small open-weights model fine-tuned for the specific workflow and run on private infrastructure, or a frontier API model running through controlled gateways. Ayoob AI builds production systems on both patterns, with the choice driven by data residency, latency, and regulatory constraints.
Related terms
Retrieval-Augmented Generation (RAG)
An architecture pattern that grounds language model outputs in retrieved documents from a private corpus, reducing hallucination and enabling answers based on the firm's own data rather than the model's training set.
Fine-Tuning
The process of further training a pre-trained language model on domain-specific data to adapt its behaviour, terminology, or output format to a particular use case or organisation.
Private AI
AI deployed on infrastructure the client controls (on-premise, in the client's cloud tenancy, or air-gapped), with no third-party LLM provider in the data path and no inference-time data export.
Want to see this technology in action?
Book a Discovery Call