Home / Gemini context window: size, limits and how to check yours
Gemini context window: size, limits and how to check yours
Gemini can hold a lot of text in view, but the amount depends on your plan, and a very long chat still degrades. Here are the numbers, the caveats, and what to do about a chat that has grown too long.
Gemini context window size by plan
Third-party summaries of Google's help pages, reviewed 2026-10-08, report these figures for the Gemini app:
| Plan | Reported context window |
|---|---|
| Free | 32K tokens |
| Google AI Plus | 128K tokens |
| Google AI Pro | 1M tokens |
| Google AI Ultra | 1M tokens |
The app limits are not the same as model or API limits, which can be larger. For a side-by-side with the other assistants see context window sizes: ChatGPT vs Claude vs Gemini.
What happens when a Gemini chat gets too long
The whole conversation is resent with each message, so as it grows the model has more to weigh. When it passes the window, the oldest parts no longer fit. You may notice forgotten instructions, repeated mistakes, contradictions and slower replies. These symptoms appear before a hard cutoff. See what context rot is and when to start a new chat.
How to check how full your Gemini chat is
The Gemini app does not show a running context meter. You can paste the conversation into the free token counter to estimate its size against each window. A real-time meter for Gemini is on the way: we are working on Gemini support for ContextMeter and will introduce it soon. It already works on ChatGPT and Claude.
What to do when it is nearly full
- Finish the current topic.
- Ask Gemini for a structured summary with the handoff prompt.
- Start a new chat and paste the summary in.
Frequently asked questions
How big is Gemini's context window?
In the Gemini app it depends on your plan. Third-party summaries of Google's help pages report 32K tokens on Free, 128K on Google AI Plus, and 1M on Google AI Pro and Ultra. Model-level and API figures can differ, so check Google's own page.
What is the Gemini context window limit?
The limit is the most text, in tokens, that the model can consider at once: your messages, its replies, and any files. When a chat fills it, older parts stop fitting and answers get worse.
Gemini context window vs Claude and ChatGPT: which is bigger?
On paid plans Gemini's 1M-token window is the largest of the three in the apps, but a bigger window does not stop long chats from degrading. See the side-by-side table on the context limits page.
Does Gemini show how much context I have used?
Not in the normal chat. Google AI Studio shows token counts for prompts, but the consumer Gemini app has no running meter.
Does ContextMeter work on Gemini?
Not yet. We are working on Gemini support and will introduce it soon. It works on ChatGPT and Claude today, and the free token counter on this site already handles any text you paste.
Last reviewed 2026-10-08. Numbers change often; see the sources linked on this page.
ContextMeter is a free, private browser extension that shows how full your ChatGPT or Claude chat is while you use it, and helps you carry the chat into a fresh one. See how it works.