Google: Gemini 3 Flash Preview

Text input Image input Video input File input Audio input Text output
Author's Description

Gemini 3 Flash Preview is a high speed, high value thinking model designed for agentic workflows, multi turn chat, and coding assistance. It delivers near Pro level reasoning and tool use performance with substantially lower latency than larger Gemini variants, making it well suited for interactive development, long running agent loops, and collaborative coding tasks. Compared to Gemini 2.5 Flash, it provides broad quality improvements across reasoning, multimodal understanding, and reliability. The model supports a 1M token context window and multimodal inputs including text, images, audio, video, and PDFs, with text output. It includes configurable reasoning via thinking levels (minimal, low, medium, high), structured output, tool use, and automatic context caching. Gemini 3 Flash Preview is optimized for users who want strong reasoning and agentic behavior without the cost or latency of full scale frontier models.

Key Specifications
Cost
$$$$
Context
1M
Released
Dec 17, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Top P Reasoning Seed Tool Choice Temperature Stop Structured Outputs Response Format Max Tokens Tools Include Reasoning
Features

This model supports the following features:

Structured Outputs Response Format Reasoning Tools
Performance Summary

Gemini 3 Flash Preview consistently performs among the fastest models, ranking in the 86th percentile across 8 benchmarks, and offers competitive pricing, placing in the 44th percentile. The model demonstrates exceptional reliability with a 100% success rate across all benchmarks, indicating minimal technical failures. Across benchmark categories, Gemini 3 Flash Preview exhibits strong performance in areas requiring precise knowledge and ethical reasoning, achieving perfect accuracy in Hallucinations, General Knowledge, and Ethics. It also shows high accuracy in Instruction Following (83.0%) and Mathematics (92.0%), with notable efficiency in these categories. While its Coding accuracy is solid at 90.0%, its Email Classification accuracy (97.0%) is competitive but not a standout. The model's reasoning capabilities are robust, scoring 88.0% accuracy. Key strengths include its speed, reliability, and high accuracy in preventing hallucinations and handling general knowledge and ethical dilemmas. Its ability to follow complex instructions and perform mathematical reasoning efficiently are also significant advantages. No notable weaknesses are apparent from the provided benchmarks.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.5
Completion $3
Input Cache Read $0.05

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Google
Google | google/gemini-3-flash-preview-20251217 1M $0.5 / 1M tokens $3 / 1M tokens
Google AI Studio
Google AI Studio | google/gemini-3-flash-preview-20251217 1M $0.5 / 1M tokens $3 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by google