|

Reference

AI Glossary

Plain-English definitions of the concepts, architectures, and techniques behind today's AI tools. 15 terms across 5 categories.

Fundamentals

Large Language Model (LLM)A neural network trained on massive text corpora that can generate, translate, summarize, and reason over natural language.TransformerThe neural network architecture behind modern AI, using self-attention to process entire sequences in parallel rather than word by word.TokenThe basic unit of text that AI models process — roughly a word fragment — that determines both input length limits and output billing.HallucinationWhen an AI model confidently generates factually incorrect or entirely fabricated information.Multimodal AIAI systems that can understand and generate content across multiple modalities — text, images, audio, and video — within a single model.

Techniques & Methods

Prompt EngineeringThe practice of crafting and optimizing input instructions to elicit better, more accurate, or more useful responses from AI models.Retrieval-Augmented Generation (RAG)A technique that supplements an LLM with relevant documents fetched at query time, grounding its answers in up-to-date, verifiable sources.Fine-TuningContinuing to train a pre-trained model on a smaller, task-specific dataset to specialize its behavior or knowledge.

Model Types

Diffusion ModelA generative model that creates images by learning to progressively reverse a noise-adding process, enabling high-quality image synthesis.

Applications

Text-to-Image GenerationAI systems that create images from natural language descriptions, enabling anyone to generate custom visuals without artistic training.AI AgentAn AI system that autonomously plans and executes multi-step tasks by calling tools, browsing the web, writing code, and taking actions in digital environments.Text-to-Speech (TTS)AI technology that converts written text into natural-sounding spoken audio, enabling voice cloning and expressive narration at scale.Text-to-Video GenerationAI systems that generate video clips from text descriptions or images, enabling automated video production at scale.

Infrastructure

Vector DatabaseA database optimized for storing and searching high-dimensional vector embeddings using approximate nearest-neighbor algorithms.EmbeddingA numerical vector representation of text, images, or other data that captures semantic meaning, enabling similarity comparisons in high-dimensional space.