> Gemini 3 Flash is a closed AI model from Google, released 17 Dec 2025. Index score 158 (GPT-4 = 100), rank 43 of 229 ranked models. API price $0.50 in / $3.0 out per million tokens, 1.05M context.

[Models](https://themodelindex.org/models/) /[Google](https://themodelindex.org/labs/google/)

# Gemini 3 Flash

Google

Released 17 Dec 2025

Closed

Small / fast

Date verified

[Compare](https://themodelindex.org/compare/gemini-3-flash-vs-deepseek-v3-2/) [Announcement](https://blog.google/products-and-platforms/products/gemini/gemini-3-flash/)

158147–169

Index score

#43 of 229 models

5/ 8

Core tests taken

4 of 6 areas · plus 16 anchor tests

−4

Versus the frontier at release

best then: GPT-5.2

Replaced by [Gemini 3.5 Flash](https://themodelindex.org/models/gemini-3-5-flash/)

$0.50/ $3.0

Price per million tokens

input / output, list price

1.05M

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

Gemini 3 Pro reached this score 1 month earlier.

## Test results · by area · tick = best by any model

Show 14 anchor tests

Reasoning & knowledge

evidence

GPQA Diamond

89%

Humanity's Last Exam

no result

Maths

evidence

FrontierMath T1-3

51%

Coding

evidence

SWE-bench Verified

75%

Terminal-Bench

64%

Agents

evidence

APEX-Agents

no result

Novel problems

evidence

ARC-AGI-2

34%

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

The Model Index

independent tests, our method

#43

of 219 · 158

LMArena · Creative writing

creative writing

#26

of 386 · 1456

LMArena · Text

overall chat quality, judged by people

#38

of 388 · 1473

LMArena · Maths

maths questions in chat

#41

of 373 · 1474

LMArena · Coding

coding questions in chat

#60

of 383 · 1505

Epoch Capabilities Index

overall capability from many benchmarks

#43

of 239 · 151.8

LMArena · Vision

understanding images

#30

of 149 · 1273

LMArena · WebDev

building web apps and front ends

#64

of 123 · 1439

LMArena · Documents

working with long documents

#38

of 41 · 1413

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

2nd

SWE-bench Verified

Epoch AI · independent

3rd

Terminal-Bench

Epoch AI · independent

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Where to use it · prices checked 1 Oct 2026

#### From the lab

Gemini

app · subscription

↗

Google AI Studio

API

↗

#### API providers · 2 · per million tokens, in / out

G

Google

$0.90 / $5.4

cheapest

G

Google AI Studio

$0.90 / $5.4

↗

Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps.

## Facts and sources

Released

17 Dec 2025

Announcement

[blog.google ↗](https://blog.google/products-and-platforms/products/gemini/gemini-3-flash/)

Release entry

Gemini 3 Flash

Scored as

Gemini 3 Flash

Artificial Analysis index

–

Output speed

–

Hugging Face

closed weights

## Google releases · around this one

Gemini 2.5 Flash-Lite

Closed

17 JUN 2025

index 119

Gemini 3 Pro

Closed

18 NOV 2025

index 161

Gemini 3 Flash

Closed

17 DEC 2025

index 158

Gemini 3.1 Pro

Closed

19 FEB 2026

index 166

Gemini 3.1 Flash-Lite

Closed

3 MAR 2026

index 141

---
Source: https://themodelindex.org/models/gemini-3-flash/ · The Model Index · data as of 2026-10-01