gemini · google · chat & research
Gemini, at Flash speed, beside everything else.
Google's models for fast answers, very large files and image edits, without a Google AI Pro plan.
- made by
- versions
- 5
- from
- 4 tokens
- included on
- Starter+
best at
What Gemini is good at
Gemini's strength is throughput. The Flash models answer quickly and cheaply, which makes them the natural choice for the steady stream of small jobs: summarise this thread, reformat this table, give me five subject lines. Gemini 3.1 Pro is the deliberate one, slower but more thorough on analysis and harder questions.
Gemini is also built to take in a lot at once. Attach a long PDF, a slide deck, a spreadsheet or an audio file and ask across all of it. On the LazySusan API, Gemini models accept prompts of up to a million tokens, which in practice means you rarely have to cut a document down before asking about it.
Two switches in the composer extend it further. Search grounds the answer in Google Search results when you need current information, and Image creates a picture with Nano Banana in the same conversation, so a product description and a first visual for it can come out of one thread.
versions
Gemini versions on LazySusan
| Version | Use it for |
|---|---|
| Gemini Auto | Picks for you: 3.5 Flash-Lite for short questions, 3.8 Flash for longer prompts and anything with files. |
| Gemini 3.1 Pro | Google's most powerful model, for analysis and questions that reward a slower, more careful answer. |
| Gemini 3.8 Flash | Fast and capable. The best balance of speed, quality and cost in the Gemini line. |
| Gemini 3.6 Flash | The previous Flash, for anyone who wants consistency with earlier work. |
| Gemini 3.5 Flash-Lite | The fastest and cheapest Gemini, made for short, simple requests. |
pricing
What Gemini costs on LazySusan
| Generation | Tokens |
|---|---|
| Gemini 3.5 Flash-Lite2–10 from a short to a long reply | 4 |
| Gemini 3.8 Flash3–15 from a short to a long reply | 6 |
| Gemini 3.1 Pro8–47 from a short to a long reply | 18 |
| Image from the chat (Nano Banana 2.1, 1K) | 41 |
Chat costs are for a message of about 300 characters; billing follows what each reply actually uses. A typical Gemini 3.8 Flash reply is 6 tokens, which makes Gemini one of the cheapest ways to get a capable answer on LazySusan.
Every plan includes Gemini. Starter is $8 a month with 1,000 tokens; unused tokens roll over while you're subscribed. Compare plans.
prompts
Prompts to try with Gemini
// big files
Here are last quarter's board pack and the finance appendix. List every figure that appears in both and flag any that don't match.
// meeting notes
This is a recording of our planning call. Give me decisions, owners and deadlines as a table, then anything left unresolved.
// current info
With Search on: what changed in the most recent update to the EU's AI Act guidance, and who does it affect?
// quick drafting
Give me ten subject lines for a re-engagement email to customers who haven't ordered in six months. Vary the tone.
// image
With Image on: a cut-away illustration of a tiny house with a loft bed, a reading nook and plants on every shelf.
tips
Getting better results
- 01Let Gemini Auto handle routine questions. It uses Flash-Lite for short prompts, at about 4 tokens a reply, and moves to 3.8 Flash when there's more to read.
- 02Save Gemini 3.1 Pro for questions where the answer will be checked or relied on. A typical Pro reply is 18 tokens.
- 03When you attach several files, say what each one is ("the first PDF is the contract, the second is the invoice"). Gemini will cross-reference them more reliably.
- 04Use Search for anything with a date attached. Gemini will ground the answer in Google results instead of relying on what it learned in training.
alternatives
When to pick a different tool
- Claude if the writing itself matters and you want a careful, well-argued answer rather than a fast one.
- Perplexity if you want each claim tied to a numbered, clickable source.
- Nano Banana if you're doing real image work and want sizes up to 4K, more shapes and editing tools.
- ChatGPT if you want structured, step-by-step reasoning on a planning or logic problem.
also on the api
Gemini in your own app
Gemini's speed and large context, through the same key and SDK as every other model.
One key, our SDKs for JavaScript and Python, prepaid credits with no subscription. LazySusan API · docs
faq
Gemini on LazySusan: questions
Yes. Gemini is included on every LazySusan plan, from Starter at $8 a month. You don't need a Google account or a Google AI Pro subscription.
Gemini 3.1 Pro, Gemini 3.8 Flash, Gemini 3.6 Flash and Gemini 3.5 Flash-Lite, plus Gemini Auto, which picks Flash-Lite for short questions and 3.8 Flash for the rest.
A typical message costs about 4 tokens on Flash-Lite, 6 tokens on 3.8 Flash and 18 tokens on 3.1 Pro. Long replies, attachments and Search add to that.
Yes. Attach documents, spreadsheets, presentations, images or audio files to a Gemini conversation and ask about them. Reading them counts toward that message's cost.
Yes. Turn on Search in the composer and Gemini grounds its answer in Google Search results.
Yes. Gemini 3.1 Pro, 3.8 Flash and 3.5 Flash-Lite are in the LazySusan developer API, alongside every other chat, image and video model, on one key and one prepaid balance.
get started
Gemini and every other model, one plan.
One login, one bill and one Library for everything you make, with the right model for each part of the job.
Starter is $8/month. Cancel anytime.