Gemini Gem: Definition and Creation (Google)

A Gemini Gem is a custom Gemini (Google) assistant, equivalent to OpenAI's Custom GPTs, configured with specific instructions and context for recurring use.

Full definition

Gemini Gems were introduced by Google in 2024-2025 as part of Gemini's ecosystem strategy, originally under Google Bard. A Gem is a preconfigured Gemini assistant that the user can create and then quickly invoke from the gemini.google.com interface or from Google Workspace apps.

A Gem is defined by:

- Behavioral instructions (role, tone, constraints) - A default context used: Drive documents, calendars, Gmail emails (depending on permissions granted) - Native integration with Google Workspace (Docs, Sheets, Slides, Calendar, Gmail) - An optional knowledge base via uploaded files

The major difference from a Custom GPT is the deep integration into Google Workspace: a Gem can natively analyze a shared Drive, read emails, write in a Doc — capabilities that in OpenAI require custom Actions to configure. On the other hand, the sharing and discovery ecosystem of Gems is less developed than the GPT Store.

Etymology

The name Gem (gem in French) evokes the notion of a precious and personalized object, and follows the lineage of Google AI product names (Gemini, before it Bard). Google explicitly positioned Gems as the cornerstone of their custom assistant strategy in 2024.

Concrete examples

Workspace productivity Gem

Create a Gem "Meeting Brief" that, from a Google Calendar event, reads the associated Drive attachments and generates a 1-page brief with context, objectives and questions to ask.

Personal assistant Gem orchestrating multiple Google apps

Configure a Gem "Inbox zero" that scans unread Gmail emails, classifies by urgency, suggests preformatted replies and creates Google Calendar tasks for important actions.

Business Gem leveraging a document corpus

Build a Gem "Financial Analyst" connected to a shared Drive containing quarterly reports from my portfolio, and answers questions by cross-referencing data.

Practical usage

To create a Gem, open gemini.google.com > Gems > Create a Gem. Define name, description, instructions, and choose default data sources (Drive, Gmail, Calendar). Gems are particularly powerful for productivity uses related to Google Workspace: email sorting, Calendar meeting summaries, leveraging a team Drive. They are invoked from the Gemini menu in any compatible Google app.

Related concepts

Custom GPTClaude SkillsGoogle WorkspaceGemini Pro

FAQ

Are Gemini Gems free?

Gems are available in the free Gemini plan, but with request limits and without access to advanced features (Deep Research, Veo). The Gemini Advanced plan (included with Google One AI Premium) unlocks full capabilities.

Can you share a Gemini Gem with your team?

Yes, since 2025 Google allows sharing Gems within an organizational Workspace. Public distribution remains more limited than the GPT Store: Google does not have an equivalent marketplace open to the general public.

What is the difference between a Gem and a Gemini agent?

A Gem is a preconfigured assistant selected manually by the user. A Gemini agent (in the 2025-2026 sense) is an autonomous system that executes multi-step tasks without human supervision. A Gem can serve as an interface to an agent, but the two concepts are not synonymous.

How to use this prompt

Copy the prompt with the button above.
Paste it into ChatGPT, Claude or your favorite AI assistant.
Replace the bracketed variables with your details, then refine the result.

About Prompt Guide

Prompt Guide is a free library of 2500+ ready-to-use prompts for ChatGPT, Claude and other AIs, with guides to learn prompting and tools to build and optimize your own prompts.

Prompt library Learn prompting Prompt builder Prompt optimizer

More definitions

Gemini Pro: Definition and Examples

Gemini Pro is a multimodal language model developed by Google DeepMind, designed to handle complex tasks of reasoning, text generation,

Grouped Query Attention: Definition and Examples

Attention mechanism that groups multiple query heads to share the same keys and values, thereby reducing memory and computational cost during inference.

Model Registry: Definition and Examples

A Model Registry is a centralized system for storing, versioning, and managing machine learning models throughout their lifecycle, from training to production deployment.

Runway ML: Definition and Examples

Runway ML is a generative AI platform specialized in creating and editing visual content (video, image, 3D) from text prompts or multimodal inputs.

Semantic Cache: Definition and Examples

A semantic cache is a caching system that stores and retrieves AI model responses based on the semantic similarity of queries, rather than exact word matches.

Thread Of Thought: Definition and Examples

Prompting technique that asks the model to unravel a continuous thread of reasoning by identifying and connecting relevant information from a long context.

Get new prompts every week

Join our newsletter.