gemini-3-flash-preview

Common Name: Gemini 3 Flash Preview

-10%On SaleReleased on Dec 18, 2025 12:00 AMKnowledge Cutoff Nov 1, 2025 12:00 AMSupportedTool Invocation

Preview of Google's next-generation Gemini 3 Flash model, optimized for speed with frontier intelligence combined with superior search and grounding capabilities.

Specifications

Context

1000K

Maximum Output

64K

Inputtext, image, video

Outputtext

Performance (7-day Average)

Collecting…

Availability Trend (24h)

Pricing

Input$0.45/MTokens

Cached Input$0.045/MTokens

Input Image$0.45/MTokens

Input Audio$0.90/MTokens

Output$2.70/MTokens

Thinking Output$2.70/MTokens

Performance Metrics (24h)

Similar Models

Gemini 2.5 Flash

$0.27/$2.25/M-10%

ctx1.0Mmax66Kavail—tps—

InOutCap

Google's most efficient workhorse model designed for speed and low-cost. Improved across key benchmarks for reasoning, multimodality, code and long context while being 20-30% more efficient.

Gemini 3.7 Flash

$0.675/$3.375/M-10%

ctx1.0Mmax64Kavail—tps—

InOutCap

RankedScience#4 RankedScicode#4

Google's most intelligent workhorse model for coding and agents, with substantial gains over 3.6 Flash in software engineering (DeepSWE v1.1 65.3%), web development (WebDev Arena Elo 1588), and knowledge-dense document processing (GDP.pdf 34.0%), at half the introductory cost.

Gemini 3.8 Flash

$0.675/$3.375/M-10%

ctx1.0Mmax64Kavail—tps—

InOutCap

RankedScience#2 RankedAgent#5

Google's workhorse model for coding and agents, arriving three weeks after 3.7 Flash with sharper multi-turn coding, rapid refactoring, code synthesis, tighter tool-execution logic, and reduced verbosity.

Gemini 3.6 Flash

$0.675/$3.375/M-10%

ctx1.0Mmax64Kavail—tps—

InOutCap

Google's workhorse model delivering better coding, knowledge work, and multimodal performance with up to 17% token reduction compared to 3.5 Flash, at lower output cost.