Google

Gemini 3 Flash Preview

by Googletext+image+file+audio+video->text6 endpoints
Neura Intelligence Index
69.0/ 100
Rank
#89
Confidence
high
Benchmarks
13(33% coverage)

Overview

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...

Capabilities

Text generation
Image understanding
Audio processing
Video understanding
Tool use / Function calling
Structured output
Extended reasoning
Prompt caching
Long context
File input
Top-P sampling
Stop sequences
Deterministic seed

Modalities

Input
TextImageFileAudioVideo
Output
Text

Technical Specifications

Context Window
1.0M tokens
Max Output
65.5K tokens
Tokenizer
Gemini
Uptime (24h)
9999.0%
Country
🇺🇸US
Release Date
December 17, 20259mo

Supported Parameters

include_reasoningmax_tokensreasoningreasoning_effortresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_p

Benchmark Performance

general

BenchmarkScoreSource
MMMLU
91.8
verified
SimpleQA
68.7
verified
MRCR v2
22.1
verified

coding

BenchmarkScoreSource
SWE-bench Verified
78
verified

math

BenchmarkScoreSource
AIME 2025
99.7
verified

reasoning

BenchmarkScoreSource
GPQA Diamond
90.4
verified
Humanity's Last Exam
43.5
verified
ARC-AGI v2
33.6
verified

multimodal

BenchmarkScoreSource
MMMU-Pro
81.2
verified
CharXiv Reasoning
80.3
verified
ScreenSpot Pro
69.1
verified

agent

BenchmarkScoreSource
MCP Atlas
57.4
verified
Toolathlon
49.4
verified

Performance Over Time

Pricing

Per 1M tokens
Input$0.50
Output$3.00
Cache read$0.05
Cache write$0.08
Audio$1.00
Reasoning$3.00
Image$0.50
Added
December 17, 2025
Last synced 9/21/2026