Generative AI model

Artificial intelligence systems that create new content, such as texts, speech outputs, images, videos, or music, from inputs using generative models and stored training data. These models learn patterns and structures from previously entered training data and generate new data with similar characteristics.

GPT (Generative Pre-trained Transformer)

A language model trained on large text datasets to understand and generate human-like text. It is commonly used in chatbots to produce natural and coherent responses.

Chatbot

A chatbot is a computer-based program designed to conduct human-like conversations in natural language with users through machine learning and natural language processing.

Token

The smallest units of text processed by GPT models. They can be words, parts of words, or punctuation marks. The number of tokens affects how much text the model can process and determines the computational cost of a request.

Tokenlimit

The maximum number of tokens that a model can process in a single input and output. If the text exceeds this limit, it must either be shortened or split into multiple requests.

Large Language Model (LLM) 

AI model trained on large amounts of text data to enable human-like language processing and generation. These models can perform tasks such as text comprehension, text generation, translation, and more.

Prompt/Input

The input provided by users to an AI system in order to receive a response. It can consist of text, images, audio, video, or other media types and significantly influences the output generated by the system.

Output

The result generated by the AI system based on the prompt. It can take the form of text, images, audio, or videos, depending on the capabilities of the system.

System prompt; System message

A predefined instruction given to the AI model to control its behavior and responses to user inputs. The system prompt determines how the AI should interact and what type of responses it can provide.

Prompt Engineering

The process of designing and optimizing input prompts to effectively guide an AI model, such as a language model, to produce accurate and useful results. This involves adjusting the formulation of questions or instructions to ensure the model generates the desired responses.

API (Application Programming Interface)

An API (Application Programming Interface) is an interface that defines rules and protocols for software applications to communicate with each other. It allows developers to utilize functions or data from other applications without having to write the entire code themselves.

Hallucination

When the AI model generates false or entirely fabricated information as a response, even though it appears to be credible.