Gemini Context Window & Token Limits (2026)

Every current Google Gemini model, Gemini 3.1 Pro, Gemini 3 Flash, and Gemini 3.1 Flash-Lite, ships a 1 million token context window in 2026. Max output differs by model: Gemini 3.1 Pro produces up to 64,000 tokens per response, while Gemini 3 Flash and 3.1 Flash-Lite go slightly higher near 65,500 tokens. One million tokens is roughly 750,000 words, enough to hold a 1,500-page book, an entire mid-sized codebase, or a year of chat history in a single prompt.

This guide explains what the Gemini context window actually means, how many tokens the current models accept, what "max output" caps your responses at, and what practically fits inside 1M tokens. Figures are verified against Google's model documentation and cross-checked with our own token counter tool. The context window is the single most important spec for long-document work, so it is worth understanding before you paste a novel or a repository into Gemini.

What a Context Window Is and What It Does Not Do

A context window is the maximum number of tokens a model can consider at once, counting both your input and the model's output. When a conversation exceeds that limit, the oldest tokens fall out of the window and the model effectively forgets them. A 1 million token window means Gemini can hold about 750,000 words of combined prompt and response in working memory at one time.

What a context window does not do is give the model permanent memory. It resets between conversations, and once tokens scroll out they are gone unless you saved them. That is why exporting and searching past conversations matters: Gemini's window is large, but it is still a window. Per Google's documentation, tokens include words, sub-words, punctuation, and whitespace, which is why token counts run higher than word counts.

Gemini Token Limits Comparison Table (2026)

The table below lists the context window, max output, and API pricing for every current Gemini model, verified for 2026. Pricing is per 1 million tokens via the Gemini API and is separate from consumer Google AI subscriptions.

The key takeaway: context window is identical across the lineup at 1M tokens, so the choice between Pro, Flash, and Flash-Lite comes down to reasoning quality, speed, and cost, not how much you can fit. Flash-Lite gives you the same 1M window at one-eighth the input price of Pro. Deprecated models (Gemini 2.5 Pro, 2.5 Flash, 2.5 Flash-Lite) are not shown; use the current 3.x lineup.

What Fits Inside 1 Million Tokens

One million tokens is approximately 750,000 words, which covers almost any single document you would realistically paste into a chat. As a rule of thumb, 1,000 words is about 750 tokens for Gemini, so the arithmetic is roughly words times 0.75.

Concretely, 1M tokens holds any one of these: a 1,500-page technical manual, an entire medium-sized software repository, a full novel plus its outline and notes, several years of email threads, or dozens of research papers at once. That is why long-context work, such as summarizing a whole book or reasoning across an entire codebase, became practical with the 3.x generation. The practical constraint is usually cost and latency, not the window size.

Running long Gemini sessions and losing track of what you asked last week? AI Toolbox (formerly Gemini Toolbox) adds full-text search, folders, and a context window meter to Gemini so nothing scrolls out of reach. Trusted by 40,000+ users with a 4.6/5 Chrome Web Store rating. Install AI Toolbox free ->

Max Output vs Context Window: The Difference

The context window is the total token budget for input plus output; the max output is a separate, smaller cap on how long a single response can be. Gemini 3.1 Pro accepts 1M tokens of input but will only generate up to 64,000 tokens in one reply. Flash and Flash-Lite cap near 65,500 output tokens.

This distinction matters for generation-heavy tasks. If you ask Gemini to produce a very long document, a full codebase, or an exhaustive report, the max output ceiling, not the context window, is what limits the reply length. The workaround is to ask for the output in sections and stitch them together. For inputs, the 1M window is almost never the bottleneck for a single document.

Why Gemini Forgets and How to Keep Your History

Gemini forgets earlier parts of a long conversation when the total tokens exceed the 1 million token window, and it forgets everything between separate conversations because the window resets. The model has no persistent memory of past chats beyond what you paste back in.

To keep important conversations, you need to search and export them before they scroll away. Gemini's native tools are limited here: sidebar search covers titles and recent chats, and full export means a Google Takeout bulk dump. AI Toolbox adds full-text search across every synced message with date, role, and exact-match filters, plus per-conversation export as TXT, Markdown, JSON, or PDF, and a live context window meter so you can see how full the window is. It works on the free Gemini plan. See ai-toolbox.co/gemini-toolbox.

Limitations and What Native Gemini Handles Well

Google's 1M context window is genuinely generous, and for most single-document tasks you will never hit it. Native Gemini handles pasting a long PDF, summarizing a big document, or reasoning over a large codebase without any extension. If your workflow is one long document at a time, native Gemini is enough.

The limitation is across conversations, not within them. Once a chat ends or the window fills, native Gemini gives you no fast way to find or reuse that content. Token counts are also estimates that vary with language and formatting, so treat the 0.75 tokens-per-word ratio as a guide, not a guarantee. For exact counts, use Google's tokenizer.

Frequently Asked Questions

What is Gemini's context window in 2026?

All current Gemini models, Gemini 3.1 Pro, Gemini 3 Flash, and Gemini 3.1 Flash-Lite, have a 1 million token context window. That is roughly 750,000 words of combined input and output. It is large enough to hold a 1,500-page book, an entire codebase, or a full novel in a single prompt.

How many words is 1 million tokens?

One million tokens is approximately 750,000 words for Gemini, using a ratio of about 0.75 tokens per word. The exact count varies with word length, punctuation, and special characters. As a quick estimate, 1,000 words is around 750 tokens. Use a token counter for precise figures before pasting large inputs.

What is the maximum output length for Gemini?

Gemini 3.1 Pro generates up to 64,000 tokens per response, while Gemini 3 Flash and 3.1 Flash-Lite cap near 65,500 tokens. Max output is separate from the 1 million token context window and limits how long a single reply can be, not how much you can input. For very long outputs, request the content in sections.

Do all Gemini models have the same context window?

Yes. Gemini 3.1 Pro, Gemini 3 Flash, and Gemini 3.1 Flash-Lite all share the same 1 million token context window. They differ on reasoning quality, speed, max output, and price, not on how much context they can hold. Flash-Lite offers the same 1M window at the lowest API cost.

Why does Gemini forget earlier messages?

Gemini forgets earlier messages when a conversation exceeds its 1 million token context window; the oldest tokens drop out of working memory. It also resets between separate conversations. To preserve important chats, search and export them with AI Toolbox before they scroll out of the window.

How do I reduce token usage in Gemini?

Be concise in prompts, avoid repeating context, and break long tasks into separate conversations. Paste only the sections you need rather than entire documents when possible. AI Toolbox's Prompt Library lets you save optimized, reusable prompts so you spend fewer tokens re-explaining context each time.

Bottom Line

In 2026, Gemini's context window is a uniform 1 million tokens across Gemini 3.1 Pro, Gemini 3 Flash, and Gemini 3.1 Flash-Lite, roughly 750,000 words. Max output ranges from 64,000 to about 65,500 tokens depending on the model. The window is large, but it is still a window: it resets between chats and drops old tokens when full. AI Toolbox adds search, folders, export, and a context window meter so your Gemini history stays reachable, free to start, $9.99/month, or $99 lifetime. From the makers of AI Toolbox (formerly ChatGPT Toolbox) (40,000+ users, 4.6/5 rating).

For related reading, see Gemini pricing plans, Gemini vs Claude, and the token counter tool.

Install AI Toolbox from the Chrome Web Store ->

Last updated: July 9, 2026