Gemini 3.5 Flash AI Model to process images, audio, video and code
Google Chat and reasoning models Upgrade to use google/gemini-3.5-flash
This page describes Google’s Gemini 3.5 Flash model and when teams pick it. Gemini 3.5 Flash is a low-latency, multimodal chat model with a 1,000,000 token context window that reads images, audio and video and supports tools like file search and computer use (preview).
What it can do
- Reads images and screenshots
- Searches the live web and cites sources
- Reads PDFs and documents
- Thinks step by step on hard questions
- Understands video input
What people use it for
Extract structured data from long documents
Upload multi-page PDFs, invoices and reports and let the model find line items, totals and dates across a million-token context. The model keeps context across long documents so you can reconcile invoices and export CSVs.
Analyze video and audio for short insights
Give a short product demo video or meeting recording and receive a concise summary, timestamps for action items, and suggested social posts. Gemini 3.5 Flash accepts video and audio inputs and returns multimodal summaries quickly.
Code review and debugging with file context
Drop a codebase or screenshots of logs and ask for fixes, explanations, or test suggestions. The model supports code execution tooling and large context windows so it can inspect multiple files and produce targeted patches.
Why it is worth it
- Handles images, audio and video as inputs for multimodal workflows.
- Maintains up to one million tokens of context for long conversations and documents.
- Fast response times suited to interactive apps and agentic tasks.
- Supports file search, URL context, and Computer Use (preview) for grounded tool use.
Questions people ask
When should I choose Gemini 3.5 Flash over smaller models?
Choose it when you need fast multimodal understanding, long-document context (up to 1,000,000 tokens), or video and audio parsing. For tiny queries or minimal cost per message, lighter models may be cheaper.
What inputs can this model read?
Gemini 3.5 Flash accepts text, images, audio and video and can use file and URL context to ground answers. It does not currently generate images or audio.
How am I billed on Katteb for using this model?
Katteb bills Gemini 3.5 Flash per message in credits; this model costs 21 credits per message on Katteb. Charges reflect actual token usage and model selection. (No dollar prices shown here.)
