> Gemini 4 Argon is a closed AI model from Google, released 30 Sep 2026. Estimated index score ~188 (GPT-4 = 100), not yet independently tested.

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

# Gemini 4 Argon

Google

Released 30 Sep 2026

Closed

Flagship

[Compare](https://themodelindex.org/compare/gemini-4-argon-vs-kimi-k3/) [Announcement](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/)

~188176–199

Estimated index score

expected range · not ranked until independently tested

–

Core tests taken

no independent results yet

−12

Versus the frontier at release

best then: Claude Opus 5.5

–

Price per million tokens

no public API price found

–

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_

Dashed ring: an estimate until Epoch AI publishes results for this model.

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

Reasoning & knowledge

no evidence yet

GPQA Diamond

no result

Humanity's Last Exam

no result

Maths

no evidence yet

FrontierMath T1-3

no result

Coding

no evidence yet

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

no evidence yet

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

LMArena · Text

overall chat quality, judged by people

#1

of 388 · 1525 · early

LMArena · Creative writing

creative writing

#1

of 386 · 1522 · early

LMArena · Coding

coding questions in chat

#1

of 383 · 1559 · early

LMArena · Maths

maths questions in chat

#1

of 373 · 1528 · early

Artificial Analysis Intelligence Index

overall capability across ten evaluations

#5

of 207 · 53

LMArena · WebDev

building web apps and front ends

#8

of 123 · 1679 · early

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

1st

coding questions in chat

LMArena · Coding

1st

maths questions in chat

LMArena · Maths

1st

creative writing

LMArena · Creative writing

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

↗

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

30 Sep 2026

Announcement

[blog.google ↗](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/)

Release entry

Gemini 4 Argon

Scored as

–

Artificial Analysis index

53 · high

Output speed

–

Hugging Face

closed weights

## Google releases · around this one

Gemini 3.7 Flash

Closed

13 AUG 2026

index 172

Gemini 3.8 Flash

Closed

2 SEP 2026

index 171

Gemini 4 Argon

Closed

30 SEP 2026

~188 est.

---
Source: https://themodelindex.org/models/gemini-4-argon/ · The Model Index · data as of 2026-10-01