Text model

Gemini by Google: a million tokens and PDF reading

Q: Which Gemini models are available in GPTunneL?

The current list is always visible in the «Every Gemini model» block above — it is built straight from the platform catalogue. Usually that means the recent Flash and Pro models plus the open Gemma ones.

Q: Which model should I use for work?

For most tasks Flash is enough — it is fast and very cheap, and its context window is the same as the larger models'. Pro is worth taking where deep reasoning matters more than speed.

Q: How is Gemma different from Gemini?

Gemma models come from the same family but with open weights: they can be downloaded and run on your own server. In GPTunneL they are available alongside the rest if you want to try them without your own infrastructure.

Q: Which languages does Gemini handle?

The models work comfortably in dozens of languages — they understand the request and answer in the same language, including long texts and documents.

Q: Can I send PDFs and images?

Yes, and it is the family's strong suit. Gemini reads scans, tables and page layout carefully — attach an invoice, a contract or a deck and work through it in plain text.

Google Gemini models in GPTunneL: 1M token context, PDF and image understanding, open Gemma weights. Pay per token, no subscription and no VPN.

Try Gemini Available models

What the Gemini family is good at

Traits Google keeps from one generation to the next — you can count on them whichever model you pick.

A million tokens everywhere

The million-token window belongs not only to the larger models but to the cheapest ones too — long documents can be read at any tier.

Document understanding

The family was trained on PDFs from the start: tables, scans and page layout are read more carefully than by most competitors.

Flash speed

The Flash models answer almost instantly and cost next to nothing — they are the choice for a stream of repetitive requests and for processing large batches.

Open Gemma models

The family has open-weight relatives you can run yourself — and try in GPTunneL without a server of your own.

Which model to pick

The family has three tiers — for different jobs and budgets.

	Flash-LiteVolume and speed	FlashThe default pick	ProHard tasks
Best for	Classification, tagging, simple extraction from text	Everyday tasks, writing, reading documents	Hard reasoning, analysis, long research
Context	1M tokens	1M tokens	1M tokens
Speed	Highest	High	Medium
Price	Lowest	Low	Above average

Figures come from the platform catalogue. Exact per-model prices are on the Pricing

Every Gemini model

Release dates follow Google's official announcements.

July 2026

Gemini 3.6 Flash Gemini 3.5 Flash-Lite

The current fast model in the family. A lighter version of the previous generation shipped alongside it.

May 2026

Gemini 3.5 Flash

The generation where the fast model learned to search the web and draw pictures in the same conversation.

April 2026

Gemma 4 31B Gemma 4 26B

Open models from the same family — the weights are published and you can run them yourself.

February — March 2026

Gemini 3.1 Pro Gemini 3.1 Flash-Lite

The larger Pro and the lighter Flash-Lite — both with a million-token window.

November — December 2025

Gemini 3 FlashGemini 3 ProGemini 3 Deep Think

The generation that put the family among the market leaders. Pro arrived in the autumn, the fast Flash by winter.

Full release history

2025

Gemini 2.5 Pro Gemini 2.5 Flash

The first generation with reasoning and PDF understanding — the traits the family became known for.

Early 2025

Gemini 2.0 FlashGemini 2.0 Pro

The generation that introduced the Flash branch — fast, cheap models for a stream of requests.

2024

Gemini 1.5 ProGemini 1.5 Flash

Context grew to a million tokens for the first time — an industry record back then.

December 2023

Gemini 1.0

The first model in the family, and Google's entry into the race with OpenAI.

How billing works

There is no subscription: you top up one balance and spend it on any model on the platform.

Pay per token

You are charged for exactly what the request and the answer used. Not using it costs nothing: no limits, no monthly fee.

One balance for every model

Gemini, ChatGPT, Claude, image and video generation — all out of the same wallet. No separate subscription per service.

Top up the way you prefer

International cards, Apple Pay, Google Pay or crypto. The minimum top-up is $5.

See pricing

How to get started with Gemini

One account for every model on the platform.

Emailyou@example.comPassword••••••••

Pull the line items and amounts out of this invoice and lay them out as a table

Invoice.pdf

Gemini 3.6 Flash

Invoice line items

The invoice has nine line items totalling 148,200 including tax.

Two items give no quantity, so their amounts are taken as they stand.

The date in the header and the date by the signature differ by a day — worth checking with the supplier.

The model switches right inside the input — you can change it mid-conversation.

The answer lands in the chat and the context is kept until the conversation ends.

Try Gemini in GPTunneL

Signing up takes a minute, and $5 on the balance is enough to see whether the model fits your task.

Try Gemini

Frequently asked questions

/01Do I need a VPN to use Gemini through GPTunneL?

No. Requests go through GPTunneL's infrastructure, so Google models open like any ordinary website — no VPN and no proxy. You do not need a Google account either.

/02Which Gemini models are available in GPTunneL?

/03Which model should I use for work?

/04How is Gemma different from Gemini?

/05Which languages does Gemini handle?

/06Can I send PDFs and images?

Gemini is Google's family of neural networks and one of the main rivals to ChatGPT. The models answer questions, write and edit text, read documents and program — and the family is strongest of all with files: PDFs, scans, tables and decks are read more carefully than by most competitors.

The million-token window belongs to the whole family, including the cheapest models — about 500 thousand words, which means a folder of documents or a long report in one go. The fast Flash models answer almost instantly and cost next to nothing, which is why they are used for streams of repetitive requests, while the larger Pro models handle hard reasoning and analysis. The family also includes the open Gemma models, which can be run on your own server.

Gemini and Gemma models are available in GPTunneL — worldwide, without a VPN and without a Google account, with one balance and billing for the tokens you actually spend.