> ## Documentation Index
> Fetch the complete documentation index at: https://docs.praxis-ai.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Glossary

> Key terms and definitions for the Praxis AI platform

<Note>
  Can't find a term? Contact our team at [humans@praxis-ai.com](mailto:humans@praxis-ai.com) to suggest additions to this glossary.
</Note>

## Key Definitions

<CardGroup cols={2}>
  <Card title="Pria" icon="robot">
    The flagship Digital Twin created by Praxis AI. Pria serves as both a demonstration of platform capabilities and a practical AI assistant designed to teach prompt engineering skills.
  </Card>

  <Card title="Praxis AI" icon="building">
    A human-first AI platform for creating, managing, and using Digital Twins, Assistants, knowledge sources, and integrations. It helps organizations scale expertise, personalize interactions, and support AI-powered workflows.
  </Card>

  <Card title="Digital Twin" icon="user">
    An AI representation of a person, expert, brand, or role, configured with specific knowledge, personality, tools, and instructions. Digital Twins help users interact with expertise in a personalized and scalable way.
  </Card>

  <Card title="Assistant" icon="wand-magic-sparkles">
    A specialized workflow inside a Digital Twin that performs a defined task or process. Assistants can follow step-by-step instructions, use approved sources, handle errors, and produce consistent outputs.
  </Card>

  <Card title="Files" icon="file">
    Documents or media uploaded to Praxis AI so they can be stored, searched, analyzed, or referenced during conversations. They provide context and source material for more accurate responses.
  </Card>

  <Card title="Memory" icon="brain">
    Saved information that helps personalize future interactions, such as user preferences, relevant context, or recurring details. Memory allows Praxis AI to provide more consistent and context-aware assistance over time.
  </Card>

  <Card title="RAG" icon="magnifying-glass">
    Retrieval-Augmented Generation — a method that lets Praxis AI search approved knowledge sources and use relevant content when generating a response, grounding answers in selected files or institutional knowledge instead of relying only on general AI knowledge.
  </Card>

  <Card title="Credits" icon="coins">
    Usage units that track access to AI features, tools, and platform activity. Credits help users and administrators monitor and manage consumption across Digital Twins and services.
  </Card>

  <Card title="Workspace" icon="folder-open">
    The user's operating environment within Praxis AI, where they access and manage conversations, Digital Twins, files, Assistants, settings, and related resources — organized by user, team, or use case.
  </Card>

  <Card title="Integrations" icon="plug">
    Connections between Praxis AI and external systems, tools, or services so information and workflows can move between platforms — accessing connected data, automating actions, or extending a Digital Twin's capabilities.
  </Card>
</CardGroup>

***

## A

<AccordionGroup>
  <Accordion title="Account" icon="building">
    The top-level organizational entity in Praxis AI. An account can contain multiple Digital Twins (instances), manage billing, and oversee users across the organization. Accounts enable centralized credit management and administrative controls.
  </Accordion>

  <Accordion title="Account Manager" icon="user-tie">
    A role that provides scoped access to manage multiple Digital Twin instances within an Account. Account Manager access is configured by the Praxis AI team. Contact [humans@praxis-ai.com](mailto:humans@praxis-ai.com) to request this role.
  </Accordion>

  <Accordion title="Act As (Impersonation)" icon="user-secret">
    An admin tool for reproducing user-reported issues. An administrator with the right permissions can temporarily act as a user after stating a reason; the session is fully audited, capped at one hour, and the rest of the admin console is locked until the admin clicks Stop. See [Memory Parameters & Act As](/mdx/admin-guide/memory-and-impersonation).
  </Accordion>

  <Accordion title="Agent Workspace" icon="desktop">
    A private, isolated cloud desktop that can be activated for autonomous agent work when enabled for your institution — its own home directory, configurable CPU, memory, and storage, and an in-browser desktop view. Idle workspaces go to sleep with files preserved and wake on demand.
  </Accordion>

  <Accordion title="AI Tool" icon="wrench">
    A specialized capability that extends a Digital Twin's functionality. Tools can perform specific tasks like searching Canvas, sending emails, generating images, executing API calls, or connecting to external services. They are configured per instance by administrators.
  </Accordion>

  <Accordion title="Assistant" icon="wand-magic-sparkles">
    A reusable prompt template that guides how a Digital Twin responds to specific types of requests. Assistants can be personal (user-created), instance-level (shared within a Digital Twin), or account-level (shared across all twins). They automate workflows and ensure consistent, specialized responses.
  </Accordion>

  <Accordion title="Audio Notes" icon="microphone-lines">
    Voice memos recorded from the Pria toolbar. Record one or more segments, save, and Pria transcribes them in the background into an **Audio Notes** collection in your personal vault, replacing the filename with a short AI-generated title. See [Audio Notes](/mdx/user-guide/files-documents/audio-notes).
  </Accordion>
</AccordionGroup>

***

## B

<AccordionGroup>
  <Accordion title="BYOK (Bring Your Own Keys)" icon="key">
    An enterprise pricing option where organizations supply their own API keys for AI providers (OpenAI, Anthropic, Google, etc.). BYOK customers receive a 40% discount on standard pricing while maintaining direct control over AI usage costs through their existing provider agreements. See [Plans and Credits](/mdx/introduction/plans-and-credits) for details.
  </Accordion>

  <Accordion title="BYOT (Bring Your Own Token)" icon="plane-departure">
    An advanced enterprise feature that lets organizations connect their own API keys for multiple AI providers — OpenAI, Anthropic, AWS Bedrock, Google, Mistral, xAI, and Stability AI — or point Pria at custom endpoints (Azure OpenAI, OpenRouter, RunPod, LiteLLM proxies, self-hosted vLLM). Pria's [Neural Engine](#n) then routes each request to the model your configuration selects. See [Switch to Praxis-AI Middleware](/mdx/introduction/byot) for setup details.
  </Accordion>
</AccordionGroup>

***

## C

<AccordionGroup>
  <Accordion title="Canvas Integration" icon="graduation-cap">
    The connection between Praxis AI and Instructure Canvas LMS. Available via [LTI 1.3](/mdx/integrations/canvas/lti-1-3) protocol or [Theme injection](/mdx/integrations/canvas/theme), allowing Digital Twins to appear directly within Canvas courses for seamless student and faculty access. Supports two-tier OAuth — a Student Access Token (read-only course data) and a Faculty Access Token (read/write for assignments, announcements, and grades).
  </Accordion>

  <Accordion title="Chat Completions API" icon="comments">
    An OpenAI-compatible REST endpoint that lets external applications interact with Digital Twins using the standard OpenAI SDK. Point any OpenAI-based app to Pria's endpoint, use your Digital Twin's Public ID as the model, and authenticate with a JWT — no other code changes required. See the [Chat Completions](/mdx/integrations/chat-completions/introduction) guide.
  </Accordion>

  <Accordion title="Collections" icon="folder-tree">
    Folder-like groupings for organizing files in your vault, with nesting, drag-and-drop, accent colors, and bulk actions (include/exclude from retrieval, reprocess, download as ZIP). See [Organizing Files into Collections](/mdx/user-guide/files-documents/collections).
  </Accordion>

  <Accordion title="Connector" icon="plug">
    A configuration that links a Digital Twin to external services or MCP (Model Context Protocol) servers. Connectors enable tools to interact with third-party services and data sources.
  </Accordion>

  <Accordion title="Conversation" icon="comments">
    A session of back-and-forth dialogue between a user and a Digital Twin. Conversations maintain context and history, can be named for easy reference, and are stored for future retrieval.
  </Accordion>

  <Accordion title="Convo Mode" icon="microphone">
    Real-time speech-to-speech conversation with a Digital Twin, in many languages. Powered by four voice providers: **OpenAI GPT-Realtime** (low-latency, client-side voice selection, full tool calling and MCP support), **Gemini Live** (Google's native multimodal audio with real-time tool calling and noise reduction), **xAI Voice Agent** (Grok-powered voice with tool calling and 5 built-in voices), or **ElevenLabs** (custom voice clones, dashboard-managed voices). A Digital Twin can also appear as a talking avatar during Convo Mode. The voice provider is configured per-instance in [Configuration](/mdx/admin-guide/configuration). See [Convo Mode](/mdx/user-guide/interface/convo-speech) for usage details.
  </Accordion>

  <Accordion title="Credits" icon="coins">
    The currency that powers AI interactions on the Praxis platform. Credits are consumed with each request based on the tokens processed (roughly 1 credit per 10,000 tokens, with a 1-credit minimum per call) and **never expire**. Credits can be pooled at the instance level or managed individually per user. Models that support caching (Claude, GPT-5) can reduce credit consumption by reusing previously processed context — savings are shown in the Dialogue Report Card. See [Plans and Credits](/mdx/introduction/plans-and-credits) for pricing and [Credit Optimization](/mdx/user-guide/credits-optimization) for saving tips.
  </Accordion>
</AccordionGroup>

***

## D

<AccordionGroup>
  <Accordion title="Digital Twin" icon="user-astronaut">
    An AI-powered virtual assistant with a unique personality, knowledge base, voice, and appearance. Digital Twins are the core product of Praxis AI—intelligent consultants available 24/7 to answer questions, perform tasks, and provide personalized support. Also referred to as an "Instance."
  </Accordion>
</AccordionGroup>

***

## E

<AccordionGroup>
  <Accordion title="Embeddings" icon="vector-square">
    Numerical representations of text that capture semantic meaning. When documents are uploaded to the IP Vault, they are converted into embeddings to enable intelligent vector search and RAG (Retrieval-Augmented Generation).
  </Accordion>

  <Accordion title="ElevenLabs" icon="microphone-lines">
    A third-party voice AI provider integrated with Praxis AI as an alternative real-time conversation provider. ElevenLabs supports custom voice clones and branded voice experiences for Digital Twins in [Convo Mode](/mdx/user-guide/interface/convo-speech). Selected per-institution via the Convo Voice Provider setting.
  </Accordion>

  <Accordion title="Enrollment" icon="user-plus">
    The process of adding users to a Digital Twin instance. Users can be enrolled individually, via CSV bulk import, or automatically through Canvas, Google Workspace, or API integrations. When integrated with Canvas or other LMS platforms, users are auto-enrolled on first access. See the [Users](/mdx/admin-guide/users) admin guide for enrollment workflows.
  </Accordion>

  <Accordion title="Entitlements" icon="key">
    Permissions that define what actions a user can perform within a specific Digital Twin instance. Entitlements control access to features like file uploads, assistant creation, admin settings, and more. See [Entitlements](/mdx/admin-guide/entitlements) for the full permission reference.
  </Accordion>

  <Accordion title="Extended Context" icon="ruler-combined">
    A premium 1-million-token context window available for select Claude models (Opus 4.6, Sonnet 4.5, Sonnet 4). The standard context window is 200K tokens. Extended context allows longer conversations and larger document processing but incurs premium pricing from Anthropic. Configured per AI model in [Personalization and AI Models](/mdx/admin-guide/personalization).
  </Accordion>
</AccordionGroup>

***

## G

<AccordionGroup>
  <Accordion title="Gallery" icon="table-cells">
    The central hub where users view and manage all their Digital Twins. From the Gallery, users can create new twins, switch between twins, start conversations, invite users, and access settings. See the [Digital Expert Gallery](/mdx/user-guide/interface/digital-expert-gallery) guide.
  </Accordion>

  <Accordion title="Gepetto" icon="wand-magic-sparkles">
    The Digital Twin Creator assistant. Gepetto helps users design and configure new Digital Twins by guiding them through creating prompts, personalities, and visual identities. Learn how to use Gepetto in the [Create Your Digital Twin](/mdx/user-guide/create-digital-twin) guide.
  </Accordion>

  <Accordion title="Gemini Live" icon="signal">
    Google's native multimodal audio streaming technology for real-time voice conversations. Supports 30 built-in voices (Puck, Charon, Kore, Fenrir, Aoede, and more), real-time tool calling, noise reduction, and transcription. Uses WebSocket-based streaming with ephemeral token security. Available as a voice provider option in [Convo Mode](/mdx/user-guide/interface/convo-speech).
  </Accordion>
</AccordionGroup>

***

## I

<AccordionGroup>
  <Accordion title="Incident (Praxis Shield)" icon="triangle-exclamation">
    A record created when [Praxis Shield](#p) flags potentially concerning activity in an institution's conversations. Each incident carries a severity, status, and categories, plus a detail report with reviewer notes and an action history. Admins triage incidents — including marking false positives — and nothing is ever enforced automatically.
  </Accordion>

  <Accordion title="Instance" icon="cube">
    A deployed Digital Twin environment with its own configuration, users, knowledge base, and settings. "Instance" and "Digital Twin" are often used interchangeably. Each instance has a unique Public ID for API integrations.
  </Accordion>

  <Accordion title="IP Vault" icon="vault">
    The secure document storage system where users upload files to train their Digital Twin. Documents in the IP Vault are processed, embedded, and made searchable via RAG, enabling the twin to provide contextually accurate answers based on proprietary content.
  </Accordion>
</AccordionGroup>

***

## K

<AccordionGroup>
  <Accordion title="KAG (Knowledge-Augmented Generation)" icon="diagram-project">
    A retrieval method that walks a knowledge graph of entities and relationships extracted from your documents at ingest time, and fuses those results with vector-based RAG via Reciprocal Rank Fusion. KAG is **always an augmentation on top of RAG** — the dense vector leg runs unconditionally; KAG adds a second graph-aware leg that recovers chunks the dense leg can miss when phrasing differs from the source text. There is no "KAG without RAG" mode. Enable per turn by choosing **RAG + KAG Fusion** in the Knowledge action menu or the **Knowledge** tab of *Your Profile*.
  </Accordion>

  <Accordion title="Knowledge Modes" icon="sliders">
    The two-axis selector that controls how Pria retrieves vault content before answering. Axis 1 is **Retrieval Mode**: *Disabled* (no retrieval, LLM only), *RAG Only* (dense vector retrieval), or *RAG + KAG Fusion* (dense + knowledge graph fused via RRF). Axis 2 is the orthogonal **Search Only** toggle: when on, Pria returns the raw vault chunks instead of an LLM-rewritten answer (no effect when Disabled is selected). See *Your Profile → Knowledge* and the \[+] action menu's Knowledge submenu.
  </Accordion>
</AccordionGroup>

***

## L

<AccordionGroup>
  <Accordion title="LMS (Learning Management System)" icon="school">
    Educational platforms like Canvas, Moodle, Blackboard, and D2L Brightspace. Praxis AI integrates with LMS platforms to provide in-context AI assistance for students and faculty.
  </Accordion>

  <Accordion title="LTI (Learning Tools Interoperability)" icon="link">
    An industry-standard protocol for integrating external tools into LMS platforms. Praxis AI supports LTI 1.3 for secure, seamless Canvas and other LMS integrations.
  </Accordion>

  <Accordion title="LXP (Learning Experience Platform)" icon="compass">
    The Praxis Learning Experience Platform — a learner-centric system for delivering personalized, adaptive learning experiences. Features include curated learning paths organized in tracks and lessons, progress tracking, leaderboards, badges, and integrated tools (WordPress, GitLab, Discord). Digital Twins are embedded directly within the LXP to provide AI-powered assistance. See [Navigating the LXP](/mdx/lxp/introduction).
  </Accordion>
</AccordionGroup>

***

## M

<AccordionGroup>
  <Accordion title="MCP (Model Context Protocol)" icon="brain">
    An open protocol for connecting AI models to external tools and data sources. Pria implements MCP in two ways: as a **server** (exposing Digital Twin capabilities to Claude Desktop, Cursor, or any MCP-compatible client) and as a **client** (connecting to remote MCP servers via [Connectors](/mdx/admin-guide/connectors)). MCP tools also work in real-time voice conversations when using the OpenAI GPT-Realtime voice provider. See [MCP Server](/mdx/integrations/mcp/introduction) for setup.
  </Accordion>

  <Accordion title="MFA (Multi-Factor Authentication)" icon="shield-halved">
    Email-based two-step verification for Pria accounts. When MFA is enabled, signing in from a new browser/device requires both a password (or OAuth) **and** a 6-digit code emailed to the account address (5-minute validity). After a successful verification, the browser is marked as a **trusted device** and skips MFA for 7 days (configurable). Currently opt-in, with a post-login nudge for admin accounts. See [Two-Step Verification](/mdx/user-guide/profile-settings/two-step-verification) (user) and [MFA Administration](/mdx/admin-guide/mfa) (admin).
  </Accordion>

  <Accordion title="Memory" icon="memory">
    Information that a Digital Twin retains about users and conversations. There are three types:

    * **Conversation History**: Past messages in a specific conversation
    * **User Memory**: Personal preferences and facts about an individual user
    * **Shared Memory**: Information accessible across all users of a Digital Twin
  </Accordion>

  <Accordion title="Middleware" icon="server">
    The Praxis AI backend infrastructure that orchestrates Digital Twin interactions. Middleware handles authentication, routes requests to AI models, manages RAG searches, executes tools, and maintains conversation state.
  </Accordion>

  <Accordion title="Mistral AI" icon="cloud">
    A French AI company providing large language models. Pria integrates Mistral via native SDK, supporting conversation (Mistral Large, Medium, Small), vision (Pixtral Large), reasoning (Magistral), code generation (Codestral, Devstral), text-to-speech (Voxtral TTS), speech-to-text (Voxtral), and embedding models.
  </Accordion>

  <Accordion title="Moderation" icon="shield-halved">
    Automatic content monitoring that screens user messages and flags inappropriate interactions before they reach the AI model. When enabled, moderation alerts are sent to designated administrators when violations occur. For reviewing and triaging flagged activity after the fact, see [Praxis Shield](#p).
  </Accordion>
</AccordionGroup>

***

## N

<AccordionGroup>
  <Accordion title="Neural Engine" icon="microchip">
    Pria's model routing layer. The Neural Engine dispatches every AI request — conversation, summaries, image generation, embeddings, voice — to the model your configuration selects, from the assistant's pinned model down through the Digital Twin's defaults. With [BYOT (Bring Your Own Token)](#b) it routes those requests through your own provider credentials and custom endpoints. See [Switch to Praxis-AI Middleware](/mdx/introduction/byot).
  </Accordion>
</AccordionGroup>

***

## O

<AccordionGroup>
  <Accordion title="Onboarding Questions" icon="clipboard-question">
    A guided questionnaire in the admin panel that helps personalize a Digital Twin. By answering questions about communication style, expertise, and preferences, administrators generate an AI prompt that makes the twin mirror a specific person's voice and approach. See [Onboarding Questions](/mdx/admin-guide/onboarding-questions).
  </Accordion>
</AccordionGroup>

***

## P

<AccordionGroup>
  <Accordion title="Pool Credits" icon="bucket">
    A setting that allows all users of an instance to draw from a shared credit pool rather than individual allocations. When pool credits are depleted, interactions pause until credits are replenished.
  </Accordion>

  <Accordion title="Praxis Shield" icon="shield-halved">
    Pria's threat-monitoring capability for institution administrators. Conversations are scanned in the background; potentially abusive or concerning activity surfaces as [incidents](#i) in the admin Threats panel, where admins review the report, add notes, progress the status, mark false positives, and share a read-only incident report with another admin of the same institution. Scanning is invisible to end users and never blocks or alters their experience — admins decide every action. See [Praxis Shield](/mdx/admin-guide/praxis-shield).
  </Accordion>

  <Accordion title="Pria" icon="star">
    The flagship Digital Twin created by Praxis AI. Pria serves as both a demonstration of platform capabilities and a practical AI assistant designed to teach prompt engineering skills.
  </Accordion>

  <Accordion title="Prompt" icon="terminal">
    The instructions that define a Digital Twin's personality, behavior, knowledge focus, and response style. Also refers to user messages sent to a Digital Twin.
  </Accordion>

  <Accordion title="Public ID" icon="fingerprint">
    A unique identifier assigned to each Digital Twin instance. Used for API integrations and system connections. Should be kept confidential.
  </Accordion>
</AccordionGroup>

***

## R

<AccordionGroup>
  <Accordion title="RAG (Retrieval-Augmented Generation)" icon="magnifying-glass">
    A technique that enhances AI responses by retrieving relevant information from a knowledge base before generating answers. RAG grounds Digital Twin responses in your actual documents, reducing hallucinations and ensuring accuracy. In Pria, RAG runs as a **dense vector** search over the personal and instance IP Vaults. You can layer **KAG** (knowledge-graph) retrieval on top for fusion, or disable retrieval entirely — see *Knowledge Modes* in the K section.
  </Accordion>

  <Accordion title="RAG Chunk Enrichment" icon="layer-group">
    An advanced RAG feature that enriches each document chunk with contextual summaries and metadata before indexing, improving retrieval accuracy for complex documents.
  </Accordion>

  <Accordion title="Reasoning Effort" icon="brain">
    A setting that controls how much time the AI spends analyzing a problem before responding. Praxis AI supports five levels: None, Low, Medium, High, and Max. Configured at the institution level in [Configuration](/mdx/admin-guide/configuration) and optionally overridden per AI model. See [AI Models](/mdx/admin-guide/ai-models#reasoning-effort) for provider-specific details.
  </Accordion>

  <Accordion title="Role" icon="user-tag">
    A user's permission level within a Digital Twin instance. Common roles include:

    * **User**: Standard access to converse and use basic features
    * **Admin**: Full access to configure settings, manage users, and view analytics
  </Accordion>
</AccordionGroup>

***

## S

<AccordionGroup>
  <Accordion title="SDK (Software Development Kit)" icon="code">
    Pre-built code libraries for integrating Praxis AI into applications. The JavaScript SDK enables web applications to embed Digital Twins with features like messaging, UI control, and event handling.
  </Accordion>

  <Accordion title="Session" icon="clock">
    A time-bounded period of user activity with a Digital Twin. Sessions track when users are actively engaged and provide analytics on usage patterns.
  </Accordion>

  <Accordion title="Shared Memory" icon="share-nodes">
    A type of memory that stores information accessible to all users of a Digital Twin. Useful for institutional knowledge, FAQs, or data that should be consistent across all interactions.
  </Accordion>

  <Accordion title="Stability AI" icon="image">
    Stability AI is a generative-media company focused on image and audio synthesis. Pria integrates Stability AI as a dedicated media-generation provider — powering the `generate_image` and `generate_audio` tools available in both chat and convo modes (video generation routes to Amazon Nova Reel or OpenAI Sora instead). Models include Stable Image Ultra / Core, SD 3.5 Large, and Stable Audio 2 (up to 190 seconds). Configured with a platform-provided key or a per-institution API key override.
  </Accordion>
</AccordionGroup>

***

## T

<AccordionGroup>
  <Accordion title="Theme Integration" icon="palette">
    A method of embedding Praxis AI into Canvas by injecting code via the Canvas Theme Editor. This approach provides a branded, always-available Digital Twin experience without requiring LTI configuration.
  </Accordion>

  <Accordion title="Token" icon="ticket">
    A unit of text processing for AI models. Tokens represent pieces of words (roughly 4 characters each). API pricing and context limits are often measured in tokens.
  </Accordion>

  <Accordion title="Tools" icon="wrench">
    Capabilities that Digital Twins can invoke to perform actions beyond text generation. Examples include web search, image generation, code execution, Canvas operations, and memory management.
  </Accordion>

  <Accordion title="Tree of Knowledge" icon="sitemap">
    A visual map of your knowledge graph in the Files panel — browse the pillars, entities, and relationships extracted from your documents, down to the source files behind them. Available for any vault that has a knowledge graph: switch the Files display from **Files** to **Tree**.
  </Accordion>
</AccordionGroup>

***

## U

<AccordionGroup>
  <Accordion title="User Memory" icon="user-circle">
    Personal information and preferences that a Digital Twin remembers about an individual user. User memory enables personalized responses that adapt to each user's communication style and needs over time.
  </Accordion>
</AccordionGroup>

***

## V

<AccordionGroup>
  <Accordion title="Variables" icon="code">
    Dynamic placeholders in assistant prompts that are replaced with actual values at runtime. Variables enable creating flexible, reusable assistants that adapt to different contexts (e.g., `${user_name}`, `${course_id}`). See [Variables](/mdx/admin-guide/variables).
  </Accordion>

  <Accordion title="Vault" icon="vault">
    The scoped file storage system within Praxis AI. Files are organized across three scopes: **User Vault** (personal files), **Institution Vault** (shared across a Digital Twin's members), and **Account Vault** (shared across all Digital Twins in an account). Also referred to as the IP Vault. See [IP Vault](/mdx/user-guide/files-documents/ipvault).
  </Accordion>

  <Accordion title="Vault Health" icon="heart-pulse">
    An A–F quality grade for the files in your vault, with categorized follow-ups such as errors, never-used files, stale files, and files that need reprocessing. The Vault Health panel in Files doubles as a filter — click a category to see just those files — and includes a knowledge-graph coverage check. See [File Processing Status](/mdx/user-guide/files-documents/file-processing).
  </Accordion>

  <Accordion title="Vector Search" icon="arrow-right-arrow-left">
    A search method that finds semantically similar content based on meaning rather than exact keyword matches. Vector search powers the RAG system, finding relevant documents even when queries use different terminology than the source material.
  </Accordion>

  <Accordion title="Video Analysis Model" icon="film">
    The model used to describe the **visual** content of an uploaded video — the scenes and on-screen content (slides, diagrams, demos, on-screen text). The resulting scene description is fused with the video's spoken transcript into a single ingested document, so on-screen content becomes searchable in RAG and is captured in the knowledge graph. It is a separate Model Use (independent of the Conversation model), currently filtered to video-capable Gemini models (`gemini-3.5-flash`, `gemini-2.5-flash`, `gemini-2.5-pro`); the platform default is set under System Models and can be overridden per Digital Twin. Runs at low media resolution for cost control and is failure-isolated — a failure never blocks the transcript. See [AI Models](/mdx/admin-guide/ai-models#video-analysis-model).
  </Accordion>

  <Accordion title="Voice Cloning" icon="microphone-lines">
    A feature (via 11Labs integration) that allows Digital Twins to speak with custom, cloned voices. Organizations can create unique voice identities for their twins.
  </Accordion>

  <Accordion title="Voice Preview" icon="headphones">
    A feature allowing administrators to audition AI voices before selecting them. Links to OpenAI TTS Playground and Google AI Studio let admins hear voice samples directly.
  </Accordion>
</AccordionGroup>

***

## W

<AccordionGroup>
  <Accordion title="Web SDK" icon="globe">
    The integration method for embedding Digital Twins into custom web applications. The Web SDK provides authentication, UI components, and APIs for full programmatic control.
  </Accordion>
</AccordionGroup>

***

## X

<AccordionGroup>
  <Accordion title="xAI / Grok" icon="bolt">
    xAI is an AI company founded by Elon Musk. Pria integrates xAI via native API, supporting conversation (Grok-4.20, Grok-4 Fast), vision, image generation (Grok Imagine), code generation (Grok Code Fast), text-to-speech (5 voices: Eve, Ara, Rex, Sal, Leo), embeddings (grok-embedding-small), and real-time voice conversations (Grok-3 Fast via WebSocket). Configured per-institution with a dedicated xAI API key.
  </Accordion>
</AccordionGroup>

***

## Z

<AccordionGroup>
  <Accordion title="Zero Data Retention (ZDR)" icon="shield-check">
    A guarantee that the AI model provider does not store prompts or model responses after a request completes — nothing is retained on the provider's servers, used for training, or available for later review. Praxis AI middleware ships only with models that comply with ZDR. The one exception is Anthropic's Mythos-class models (e.g., Claude Fable 5), which are subject to a mandatory 30-day data retention that overrides ZDR agreements; these models are flagged with a warning in the model selector. See [Anthropic's retention notice](https://support.claude.com/en/articles/15425996-data-retention-practices-for-mythos-class-models).
  </Accordion>
</AccordionGroup>

***

<Note>
  **Can't find a term?** [Contact our team](mailto:humans@praxis-ai.com) to suggest additions to this glossary.
</Note>