Structured Outputs enforce a specific response shape, often with schemas, so AI output can be parsed and consumed reliably by software.

Reasoning Effort is a controllable depth setting for model thinking, balancing answer quality, latency, and cost.

A Context Window is the maximum amount of tokens a model can process at once, including instructions, conversation history, and retrieved data.

Tool Calling is the capability for a model to decide when to call connected tools, then use tool results to complete a task.

Retrieval-Augmented Generation (RAG) enriches model outputs by fetching external knowledge at runtime and conditioning generation on it.

Grounding is the practice of constraining generation with verifiable sources so outputs are accurate, attributable, and context-specific.

Evaluations (Evals) are benchmarked test suites used to track model behavior, regressions, and improvements against defined criteria.

Human-in-the-loop is an operating model where people validate, correct, or approve AI outputs before final execution.

Model Routing is the strategy of directing requests to the most suitable model dynamically using policy and workload rules.

Related AI terms: Partner Models and Model Card.

Text-to-Image Generation creates visuals directly from prompt instructions. Most modern systems rely on a Diffusion Model and can be steered using a Reference Image.

Style Reference lets you guide the aesthetic of generated assets by pointing the model to example visuals. It is frequently combined with Reference Image inputs in Text-to-Image Generation workflows.

Partner Models are external foundation models exposed through a single product interface. Product teams often combine them with Model Routing and governance artifacts like a Model Card.

A Model Card summarizes model behavior, scope, limitations, and known risks for operators and users. It supports safety controls such as Guardrails and helps teams reason about Hallucination risk.

A Diffusion Model creates images through iterative denoising steps conditioned on prompts and controls. It is the backbone of many Text-to-Image Generation systems and can be steered by Classifier-Free Guidance (CFG).

ControlNet augments diffusion generation with explicit structural conditions such as edges, depth, or pose. It improves control in Diffusion Model systems and often uses cues from Image Segmentation.

Vision Transformer (ViT) applies transformer attention mechanisms to image patches for classification and representation learning. It is widely used in multimodal stacks with CLIP and in segmentation systems like Segment Anything Model (SAM).

Image Segmentation partitions an image into labeled regions to isolate objects or areas for editing. It is core to Segment Anything Model (SAM) workflows and precision operations like Generative Fill.

The ability of AI models to translate intent—expressed as prose, pseudocode, or examples—into executable source code. Code generation underpins vibe coding, autocomplete tools, and agentic workflows. Generated code should always be reviewed for correctness, security, and maintainability, since models can produce plausible-looking but incorrect implementations.