What is Gemini 3.7 Flash?
Gemini 3.7 Flash is a Google reasoning model with multimodal input and text output. Here on NanoBanana Pro, you can use it in a familiar chat workspace: ask a question, add context and refine the answer through follow-up messages. It is a useful starting point for drafting, explaining code and discussing visual references.
Put Gemini 3.7 Flash to work
Understand code before changing it
Paste a small function, its expected behavior and the error you see. Ask for an explanation, a focused fix and cases to test. Review the proposal and run it in your own environment before relying on it.
Help me debug this code. First explain what it does, then identify the likely cause of the error, propose a minimal fix and list three tests. Ask me for the code and error message before starting.
Write with a clear audience and purpose
Give the model a reader, a goal and a tone. Work from an outline to a first draft, then ask for a shorter or more specific revision. Supplying your facts helps keep the result grounded in your material.
Help me write a product introduction. Ask about the product, audience and three verified benefits first. Then produce a concise headline, a 100-word introduction and a clear call to action. Do not invent claims.
Turn dense notes into decisions
Paste the relevant text or attach a PDF for text extraction. Ask for a summary, open questions and action items with references to the original material. Break very long sources into sections; the official context limit is not a promise about this chat interface.
Organize the notes I will provide into decisions, action items, owners and unanswered questions. Cite the original sentence for each decision. Mark missing owners or dates as unspecified rather than guessing.
Discuss an image or interface
Attach an image and describe what you want to understand. For an interface, ask about hierarchy, readability and navigation. For a chart, request a description and flag any labels that are too small to read.
I will attach an interface screenshot. Review its information hierarchy, readability and navigation. Separate what is visible from your assumptions, then suggest three specific improvements.
Official Gemini 3.7 Flash specifications
- Developer
- Model ID
- gemini-3.7-flash
- Official input limit
- 1,048,576 tokens
- Official output limit
- 65,536 tokens
- Official inputs
- Text, images, audio, video and PDF
- Output
- Text
Gemini 3.7 Flash vs Gemini 3.6 Flash
Google’s August 2026 model card reports the following results. Higher is better for these measures. These are provider evaluations, not measurements of this website; scores do not guarantee the quality or speed of your conversation.
| Provider benchmark | Gemini 3.7 Flash | Gemini 3.6 Flash |
|---|---|---|
| FrontierCode 1.1 Main | 43.6% | 34.4% |
| DeepSWE v1.1 | 65.3% | 48.6% |
| Code Arena · Elo | 1588 | 1538 |
| AutomationBench | 30.4% | 17.0% |
How to use Gemini 3.7 Flash online
Choose the model
This page starts with Gemini 3.7 Flash selected. The left model navigation and the chat selector stay in sync. Switching models keeps the current conversation; start a new chat when you want a clean context.
Describe the result you need
Specify your task, relevant facts, audience and desired format. You can use a prompt from this guide as a draft and edit it before sending.
Sign in and send
Sending a message requires an account with available points. Check the current price in the composer, then submit your prompt. Follow up with corrections or more context to improve the answer.
How is Gemini 3.7 Flash chat priced?
This workspace charges points based on usage. The composer reads reference pricing from the live model catalog, and completed requests show their actual points cost. Longer inputs, longer replies and conversation history can increase usage. Check the current balance and point packages before sending; third-party API dollar prices are not this site’s checkout price.
View points and plans ↗Gemini 3.7 Flash FAQ
Can I use Gemini 3.7 Flash for free?
You can browse this page and prepare a prompt without signing in. Sending requires an account with available points. Any trial allowance depends on the offer shown in your account; unlimited free chat is not promised.
Do I need a Google API key?
No. This page uses your NanoBanana Pro account and point balance. Developers who want to call Google directly should follow the official Gemini API documentation linked below.
Can Gemini 3.7 Flash read images and PDFs?
Google lists image and PDF understanding among the model’s official inputs. This workspace accepts JPEG, PNG, WebP and PDF files up to 24 MB. PDFs go through text extraction, so scanned pages or complex layouts may need a clearer image or manually pasted text. Audio and video uploads are not available.
Does it search the web or let me adjust thinking?
Those capabilities exist in Google’s API, but the current site integration does not enable web search or configurable thinking for this model. Disabled controls in the composer indicate unavailable options.
How long can an input or answer be?
Google documents a 1,048,576-token input limit and a 65,536-token output limit. This chat currently requests up to 4,096 output tokens per answer, and upload limits are separate. Divide large tasks into smaller parts and ask for a continuation when needed.
What happens when I select another model?
The model navigation and chat selector update together. Your existing messages stay in the current conversation, and the next request uses the newly selected model. A reply already being generated keeps the model that started it. Use New chat to begin a separate conversation.
Can it generate images?
Gemini 3.7 Flash produces text. You can ask it to describe an image or write an image prompt, then use an image-generation model from the model library to create the picture.
Is NanoBanana Pro the official Google Gemini website?
No. NanoBanana Pro is an independent platform providing a shared workspace for multiple AI models. Google develops Gemini; model specifications and provider evaluations are linked below.
Sources and model information
Information checked
Start with one useful question
Bring your context, choose a clear output format and refine the result one message at a time.
Chat with Gemini 3.7 Flash ↑