Quick Answer
Gemini in Google Docs can now create and edit images, diagrams, and infographics directly in your document, using your text as context. You can generate visuals, refine existing ones, or update several at once with natural language prompts. The feature rolled out from July 28, 2026, on the web, for eligible paid Workspace and Google AI plans.
Key Takeaways
- Gemini in Google Docs can now generate and edit images, diagrams, and infographics inside your document.
- The visuals are context-aware, drawing on your document’s text to stay relevant.
- Three functions cover creating visuals, refining existing ones, and editing multiple at once.
- It is web-only and limited to eligible paid Workspace and Google AI Pro or Ultra plans.
- Always review AI-generated visuals, since they can contain errors or hallucinations.
What can Gemini do with images in Google Docs now?
Gemini in Google Docs can now generate and edit images, diagrams, and infographics directly inside your document, without leaving Docs or using a separate tool. It reads the text around your cursor and uses that context to make the visual relevant, so a diagram or infographic reflects what you actually wrote. Google announced the feature in a Workspace update, on its official Workspace Updates blog.
The practical appeal is speed. Instead of writing in Docs, switching to an image tool, generating a graphic, and pasting it back, you describe what you want and Gemini creates it in place. For reports, proposals, study notes, and long documents, that removes a lot of friction from adding visuals.
It is powered by Google’s latest image generation models, and it handles more than simple pictures. You can ask for a diagram that explains a proposal, an infographic that summarizes a section, or cover art for the top of a document. The visuals are editable and can be refined with follow-up prompts.
How do you create an image or diagram in Google Docs with Gemini?
You create an image or diagram in Google Docs with Gemini by describing it in plain language from the Gemini side panel or the bottom bar. The feature is web-only for now, so you need Docs in a browser rather than the mobile app. Work through these steps.
- Open your document in Google Docs on the web, where the feature lives.
- Open the Gemini side panel, or use the bottom bar in the document.
- Describe the visual you want, such as “add a diagram explaining this proposal” or “create an infographic summarizing this section.”
- Review the generated image, then insert it into your document.
- Refine it with a follow-up prompt, for example “change the aspect ratio to 4:3” or “match the style of the other images.”
Because Gemini pulls from your document, the more specific your writing and prompt, the more relevant the result. Stop here and double-check any diagram or infographic before you rely on it, since AI can get facts, labels, or figures wrong. For the broader set of Gemini writing and formatting tools, our guide on using Gemini in Google Docs, Sheets, and Slides covers the rest.
What are the three new Gemini visual functions?
Gemini’s new visual tools in Docs break down into 3 functions: creating visuals, refining existing ones, and editing several at once. Each maps to a common task, so you are not limited to generating one image at a time. Here is how they work.
| Function | What it does | Example prompt |
|---|---|---|
| Create visuals | Generates a diagram or infographic from your text | “Create an infographic summarizing this doc” |
| Refine existing | Adjusts an image’s style, ratio, or details | “Change this image’s aspect ratio to 4:3” |
| Batch edit | Creates or updates multiple visuals at once | “Add an infographic to each key section” |
The batch function is the standout for long documents, since it can add section-specific graphics across an entire report in one request. The refine function matters too, because it means you can iterate on a visual with words rather than redoing it. Together they make visuals feel like part of writing rather than a separate chore. Before running a batch edit across many visuals, though, back up your document, or at least make a copy, since a bulk change is harder to undo than a single one.
Who can use the new Gemini image features in Docs?
The new Gemini image features in Docs are limited to eligible paid plans, not free Google accounts. Google lists availability for Business Standard and Plus, Enterprise Standard and Plus, Education Plus, and Google AI Pro and Ultra subscribers, along with the Google AI Pro for Education add-on. You also need Workspace smart features turned on.
For organizations, an admin controls access, and the tools are on by default where Gemini for Workspace in Drive is enabled. Individual users on a qualifying plan can check that smart features are enabled in their account Settings. If you are on a free personal Google account, this specific feature is not included, though Google often widens availability over time.
The rollout began on July 28, 2026, across both Rapid and Scheduled Release domains, with full visibility expected within about 15 days. So if you have a qualifying plan but do not see it yet, it may still be reaching your account. Google’s official Workspace feature drop post lists the rollout details.
What are the limitations to know?
The main limitations are that the feature is web-only, paid-only, and, like all AI, capable of mistakes. It does not work in the Google Docs mobile apps yet, so you need a browser. It is also gated behind qualifying Workspace and Google AI plans, which rules out free accounts.
The more important limit is accuracy. Gemini generates visuals from your text, but it can still misread context, mislabel a diagram, or invent details, which is the usual risk with generative AI. A chart or infographic that looks polished can still be wrong, so treat the output as a fast first draft rather than a finished, verified graphic.
Review every generated visual against your source material before sharing it, especially for anything with numbers, names, or claims. This is the same caution that applies to AI text. Used carefully, the tool saves real time, but it does not remove the need for a human check.
Is this better than using a separate AI image tool?
For document visuals, the built-in Gemini tool is usually more convenient, while a dedicated image tool still wins for standalone or high-control art. The advantage of the Docs version is context and speed: it knows what your document says and inserts the result in place, which is ideal for diagrams and infographics tied to your writing. That is hard to match by copying prompts into another app.
A separate generator earns its place when you want maximum control over style, higher-quality artwork, or images unrelated to a document. Tools focused only on image generation often offer finer settings and stronger results for creative work. The two are not really competitors so much as different jobs.
If you want to compare options, our roundup of the best free AI image generators covers standalone tools, and our guide to AI presentation makers is useful for turning documents into slides. For the assistant behind all of this, see our explainer on what Google Gemini AI is, and our look at Gemini Notebook covers another Gemini research tool worth knowing.
FAQ
Can Gemini create images in Google Docs?
Yes, Gemini can now create and edit images, diagrams, and infographics directly in Google Docs using your document’s text as context. You access it from the Gemini side panel or the bottom bar on the web.
Is the Gemini image feature in Docs free?
No, it is limited to eligible paid plans, including Business, Enterprise, Education Plus, and Google AI Pro or Ultra, with smart features enabled. Free personal Google accounts are not included.
Does it work in the Google Docs mobile app?
No, the feature is currently web-only, so you need Google Docs in a browser. Google has not announced a mobile release date.
Can Gemini edit visuals I already have in a doc?
Yes, you can refine existing visuals with prompts, such as changing an aspect ratio or matching a style. It can also update multiple images at once.
Are the AI-generated visuals always accurate?
No, generative AI can mislabel diagrams or invent details, so the visuals can contain errors. Always review them against your source material before sharing.