User Intent

GPT stands for "Generative Pre-Trained Transformer". This term encapsulates the core functionalities and architecture of an AI model . This type of AI model is used for natural language processing and is capable of generating human-like text based on extensive data training.

A record of the conversation between the user and the AI, which can be referenced for continuity.

The amount of information within a conversation the AI Assistant can retain at any given time. The size of the context window varies depending on the AI’s design and capabilities. In many AI systems, the context window is measured in tokens, which determine how much of the conversation the AI can reference at once.

Compound AI refers to an advanced AI system that combines multiple AI models, algorithms, or methodologies to create a more capable system than any single LLM could achieve alone. Like a chemical compound that forms from different elements, Compound AI integrates various AI technologies (such as machine learning, natural language processing, computer vision, and inference-level optimization) to handle complex tasks such as logic and coding, and provide stronger output. This approach allows the system to leverage the strengths of each component while potentially compensating for individual weaknesses across LLMs.

A series of interactions between a user and the AI Assistant, typically involving multiple related tasks. The AI Assistant uses context from previous exchanges to maintain coherence and respond appropriately as the chat progresses. During a chat, the user might request the assistant complete multiple tasks such as planning an itinerary or researching restaurants, all within the same thematic context.

A method of communication where the user interacts with the AI Assistant using spoken commands, typically through a device’s microphone. Voice interaction relies on speech recognition technology to interpret the user's spoken inputs and respond accordingly.

Deep research is AI-driven approach to gathering, analyzing and consolidating mass amounts of data across the web, documents, and other sources to deliver a comprehensive report. Deep Research is similar to having an AI research assistant or analyst, which uses an LLM built on AI reasoning to develop and execute a research plan for you.

A process where the user adjusts or rephrases their prompt to obtain a more accurate or relevant response from the AI Assistant. Refinement helps the AI better understand the user's specific needs, ensuring that the output aligns more closely with the desired outcome.