1. Introduction
Gemini 3.5 Flash is a ModelVia model.Gemini 3.5 Flash is Google's fast multimodal model for low-latency reasoning, coding, grounded answers, and million-token context workflows.
This page combines the model overview, interactive testing surface, request schemas, and production API workflow in one place.
2. Key Features & Innovations
- • Low-latency chat
- • Long Context
- • Coding
- • Grounded Search
3. Model Architecture & Technical Details
Gemini 3.5 Flash pairs a 1M-token input window with a 65,536-token output limit for fast multimodal reasoning, coding, and grounded responses through the Gemini API.
Describe the result you want the model to create.
Optional system-level behavior instructions.
Higher values make the response more varied.
Nucleus sampling cutoff. Lower values keep responses more focused.
Limits sampling to the most likely K tokens when supported by the route.
Maximum response length. Gemini 3.5 Flash supports up to 65,536 output tokens.
4. Performance Highlights
- • Supports up to 1,048,576 input tokens and 65,536 output tokens.
- • Standard token pricing is configured at 0.15 credits / 1K input and 0.9 credits / 1K output.
