GoogleProprietaryMultimodalVisionSmall Models

Gemini Flash

Google’s Gemini 3.8 Flash workhorse — fast, token-efficient multimodal model for agentic workflows, coding, and high-throughput apps where latency and cost matter. Succeeds 3.7 Flash (GA Aug 2026); 3.7 remains supported for efficiency-first workloads.

Vision · Audio · Tool calling · Thinking · Coding

Last reviewed: 3 September 2026

Official pricing →

Overview

Google’s Gemini 3.8 Flash workhorse — fast, token-efficient multimodal model for agentic workflows, coding, and high-throughput apps where latency and cost matter. Succeeds 3.7 Flash (GA Aug 2026); 3.7 remains supported for efficiency-first workloads.

Capabilities

  • Vision: Yes
  • Audio: Yes
  • Tool calling: Yes
  • Thinking: Yes
  • MCP: No
  • Coding: Yes
  • Structured output: Yes

Technical specifications

Provider
Google
License
Proprietary
Context window
1.0M
Release
2026-09
Modalities
Text, Image, Audio, Video
Vision
Yes
Audio
Yes
Tool calling
Yes
Thinking
Yes
MCP
No
Open weights
No
API
Yes
Pricing (input)
~$0.75 per 1M tokens (3.8 Flash intro through 2026-12-31)
Pricing (output)
~$3.75 per 1M tokens (intro); ~$1.50 / $7.50 from 2027-01-01

Supported modalities

Text · Image · Audio · Video

Context window

1.0M (1,048,576 tokens)

Pricing

Input: ~$0.75 per 1M tokens (3.8 Flash intro through 2026-12-31)
Output: ~$3.75 per 1M tokens (intro); ~$1.50 / $7.50 from 2027-01-01

Same introductory rates as 3.7 Flash; expire Dec 31, 2026. Verify Google AI / Vertex pricing; Flash-Lite is cheaper for volume. 3.8 Flash Cyber is Fairwind-only — not a public API SKU.

Availability

API: Yes
Chat UI: Yes
Open weights: No

Pin gemini-3.8-flash. Gemini API / AI Studio, Antigravity, Android Studio, Gemini Enterprise, and Google AI Pro/Ultra (Gemini app, Search AI Mode, Sheets). 3.7 Flash remains supported. 3.8 Flash Cyber is restricted to the Fairwind Program.

Use cases

  • High-volume customer support
  • Agentic tool loops and coding assistants
  • Cost-sensitive RAG generation
  • Mobile / edge-adjacent latency budgets

Strengths

  • Same Flash speed/cost as 3.7 with stronger coding, agent, and multi-step reasoning per Google’s published evals
  • Introductory $0.75/$3.75 per 1M tokens through 2026-12-31
  • Large context relative to price (~1.05M)

Limitations

  • Weaker than Pro-class on hardest reasoning (3.5 Pro still partner-testing as of 2026-09)
  • Closed weights
  • Intro token rates rise on 2027-01-01; may use more tokens at higher effort

Related guides

Related benchmarks

Related research

Related GitHub

Related tools

Related rankings

Related comparisons

Companies

Explore more models

All models →