> Gemma 3 is an open-weights AI model from Google, released 12 Mar 2025. Index score 112 (GPT-4 = 100), rank 154 of 229 ranked models. API price $0.080 in / $0.45 out per million tokens, 131K context.

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

# Gemma 3

Google

Released 12 Mar 2025

Open

Date verified

[Compare](https://themodelindex.org/compare/gemma-3-vs-gemini-2-5-pro/) [Announcement](https://blog.google/innovation-and-ai/technology/developers-tools/gemma-3/)

112100–123

Index score

#154 of 229 models

1/ 8

Core tests taken

1 of 6 areas · plus 9 anchor tests

−24

Versus the frontier at release

best then: o1

$0.080/ $0.45

Price per million tokens

input / output, list price

131K

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_

o1-mini reached this score 6 months earlier.

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

Show 7 anchor tests

Reasoning & knowledge

evidence

GPQA Diamond

48%

Humanity's Last Exam

no result

Maths

evidence

FrontierMath T1-3

no result

Coding

evidence

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

evidence

APEX-Agents

no result

Novel problems

no evidence yet

ARC-AGI-2

no result

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

#154

of 219 · 112

LMArena · Creative writing

creative writing

#160

of 386 · 1349

LMArena · Text

overall chat quality, judged by people

#175

of 388 · 1365

LMArena · Maths

maths questions in chat

#203

of 373 · 1322

Epoch Capabilities Index

overall capability from many benchmarks

#135

of 239 · 130.0

LMArena · Coding

coding questions in chat

#235

of 383 · 1358

LMArena · Vision

understanding images

#95

of 149 · 1160

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

Not in the top three among current models on any category we track.

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 · 4 · per million tokens, in / out

D

DeepInfra

$0.080 / $0.16 · fp8

cheapest

N

Novita

$0.12 / $0.20 · bf16

↗

N

Nebius

$0.10 / $0.30 · fp8

↗

P

Parasail

$0.080 / $0.45 · fp8

↗

#### Run it yourself

🤗

Official weights

google/gemma-3-27b-it

↗

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

12 Mar 2025

Announcement

[blog.google ↗](https://blog.google/innovation-and-ai/technology/developers-tools/gemma-3/)

Release entry

Gemma 3

Scored as

Gemma 3 27B

Artificial Analysis index

–

Output speed

–

Hugging Face

[google/gemma-3-27b-it ↗](https://huggingface.co/google/gemma-3-27b-it)

## Google releases · around this one

Gemini 2.0 Flash

Closed

5 FEB 2025

index 121

Gemini 2.0 Pro

Closed

5 FEB 2025

index 121

Gemma 3

Open

12 MAR 2025

index 112

Gemini 2.5 Pro

Closed

25 MAR 2025

index 141

Gemini 2.5 Flash

Closed

17 APR 2025

index 132

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