Glossary

Definitions of terms used in the curriculum.

Showing 401–440 of 902 terms in the glossary

Letter G17 terms

Gitea Actions

  • Concepts
The built‑in continuous integration/continuous deployment (CI/CD) system that runs workflow files stored in .gitea/workflows/.

GitHub integration

  • Features
A two-way sync between your Lovable project and a GitHub repository so your code is version-controlled and other developers can collaborate on it.
1 other definition

GitHub integration · Features

A v0 connection that automatically commits every code change to a dedicated branch in your GitHub repository and lets you open a pull request to merge it into main.

goal

  • Concepts
A plain-English field in agents.yaml stating what an agent is trying to achieve, steering its reasoning and output.

Google AI Pro

  • Concepts
A paid subscription tier for Gemini that gives 4× higher usage limits and a 1-million-token context window (supporting up to 1,500 pages of text).

Google AI Ultra

  • Concepts
The highest Gemini subscription tier, offering the largest usage allowances and exclusive access to features such as Deep Think and enhanced Deep Research visuals.

google/gemma-4-31b-it:free

  • Models
An example OpenRouter model ID showing the standard format: provider name, a slash, the model name, and the :free variant suffix to use it at no cost.

GPT Store

  • Features
A public directory at chatgpt.com/gpts where anyone can browse and use GPTs created by OpenAI, partners, and the wider community, organised by category.

gpt-5.3-codex-spark

  • Models
A text-only research-preview Codex model built for near-instant response (over 1,000 tokens per second), optimized for real-time coding iteration — available to ChatGPT Pro subscribers.

gpt-5.4-mini

  • Models
A faster, lower-cost Codex model suited for simpler or repetitive tasks where speed matters more than maximum reasoning power.

gpt-5.5

  • Models
OpenAI's most capable Codex model (as of mid-2026), best for complex coding, research, and multi-step tasks — the recommended default when quality matters most.

gpt-oss-120b

  • Models
OpenAI's 120-billion-parameter open-source model hosted on Cerebras hardware, capable of generating roughly 3,000 tokens per second.

GPTs

  • Features
Custom versions of ChatGPT built with specific instructions, uploaded knowledge, and selected tools for a particular purpose — for example, a GPT focused on cooking advice or legal summaries.

GPU server

  • Infrastructure
A computer equipped with graphics‑processing units that accelerate AI model calculations.

Green

  • Process
The label for an example that the current prompt gets right after a change.

Green check

  • Interface
The green check mark shown on a commit that indicates all CI workflow steps completed successfully.

grounding

  • Core
Tying an AI answer to specific sources you provide, so each claim can be traced back. A grounded answer is much easier to check than one written from the model's memory.

Also known as: grounded answer, grounded

groupby

  • Concepts
A pandas operation that splits a table into groups based on the values in a column (like splitting experiment rows by treatment condition) so you can calculate statistics for each group separately.

Letter H23 terms

hallucination

  • Core
An answer from an AI model that sounds confident but is false or made up, such as an invented source or a wrong figure. Check facts, numbers and references against the original before you rely on them.

Also known as: hallucinate, confabulation

hello-world container

  • Commands
A small test image that prints a success message when run, confirming Docker is installed correctly.

hermes --continue

  • Commands
Reopen Hermes where you left off — it reloads your last session.

hermes --tui

  • Commands
Start Hermes in its modern text UI (a full-screen terminal interface).
2 other definitions

hermes --tui · Commands

Starts Hermes in its text user interface (TUI) mode — a keyboard-driven chat panel that runs entirely inside your terminal with live streaming and modal overlays.

hermes --tui · Commands

Launches Hermes Agent in its terminal user interface mode instead of the classic command-line prompt interface.

hermes doctor

  • Commands
Health check — verifies your config, model connection and tools, and reports what is wrong.

hermes gateway setup

  • Commands
Connect chat apps (Telegram, Signal, Slack…) so you can message your agent.

hermes gateway status

  • Commands
Check whether your connected chat apps are reachable.

hermes model

  • Commands
Choose or switch the model (the "brain") Hermes runs on.

hermes sessions list

  • Commands
List your saved conversations.

hermes setup

  • Commands
First-run wizard — it asks which model provider to use and writes your config file.
2 other definitions

hermes setup · Commands

A setup command for the Hermes AI agent that guides you through configuring your model provider, including connecting it to a local LM Studio server.

hermes setup · Commands

Runs the Hermes Agent interactive setup wizard that walks you through configuring all or part of your Hermes installation.

hermes setup --portal

  • Commands
Set up Hermes using a managed Nous Portal model, so you skip picking your own provider and key.

hero section

  • Concepts
The top part of a webpage that usually contains a large image or headline, which the AI can build from a design mockup.

Hetzner Cloud

  • Provider
A German cloud service that provides virtual machines billed by the hour with data stored in the EU.

hex code

  • Settings
A six‑digit combination (like #FF5733) that specifies an exact color for branding.

high resolution setting

  • Settings
A selection in Sora 2 Pro that sets the output video’s detail level to ‘High’ for better visual quality.

high‑stability

  • Settings
An option when cloning a voice that makes the synthetic voice work reliably even with imperfect input audio.

History

  • Features
The chronological log of every change made to your project, letting you revert to any earlier version or save a checkpoint.

hook

  • Concepts
A script that fires automatically on an event (e.g. after every file edit) — for things that should happen without you asking.

horizontal video (16:9)

  • File Types
A widescreen video format wider than it is tall, typical for YouTube or presentations.

HTTP TTS API

  • Backend
A web interface that accepts text via an HTTP request and returns spoken audio, used by the plugin to produce speech.

http://<your-ip>:1234/v1

  • API & URLs
The address other devices on your local network use to reach your LM Studio server — replace with your computer's actual network IP address (e.g. 192.168.1.5).

http://localhost:1234/v1

  • API & URLs
The full address of LM Studio's OpenAI-compatible API server on your own machine — localhost means 'this computer', 1234 is the default port number, and /v1 is the path prefix for the OpenAI-compatible endpoints.

http://localhost:1337

  • Concepts
A common way to write the base URL of Jan's built-in API server; Jan's docs show http://127.0.0.1:1337 as the actual default address, and localhost is simply a hostname alias for 127.0.0.1 on most computers.