ModelBeat
Models/Gemini 4 Argon
All models
G

Google DeepMind: Gemini 4 Argon

google/gemini-4-argon

Gemini 4 Argon is Google DeepMind’s proprietary AI model, released Oct 13, 2026. It scores 167 on the Epoch Capabilities Index, costs $4/M input and $20/M output tokens, and handles a 1M-token context window. Its strongest use case is agentic & tools, where it ranks in the 99th percentile of the models Model Beat tracks.

Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents.

Context
1M
Input
$4 / 1M
Output
$20 / 1M
Released
Oct 13, 2026
Modalities
audiofileimagetextvideo→text
Strongest atAgentic & Tools top 1%Reasoning & Knowledge top 7%Coding top 8%

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where Gemini 4 Argon lands among all models tracked on Model Beat.

92.0
Intelligence Index
Epoch AI
92nd percentile of tracked models
92.0
Coding Index
Epoch AI
92nd percentile of tracked models
99.0
Agentic Index
Epoch AI
99th percentile of tracked models

Reasoning

5 evals
ARC-AGI98.5%

Abstract visual reasoning

ARC-AGI-291.3%

Harder abstract reasoning

GPQA Diamond96.4%

Graduate-level scientific reasoning

SimpleBench91.7%

Common-sense trick questions

SimpleQA Verified76.0%

Factual accuracy & hallucination

Coding

1 evals
WebDev Arena1680

Human-rated web development

Math

1 evals
AIME 2024/2025100.0%

Olympiad-qualifier math

Agentic & Tools

1 evals
APEX82.4%

Multi-step agentic tasks

In the news

12

Compare Gemini 4 Argon with

Frequently asked questions

What is Gemini 4 Argon?

Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents.

Who created Gemini 4 Argon?

Gemini 4 Argon is an AI model developed by Google DeepMind, part of the Gemini family, released Oct 13, 2026.

How much does Gemini 4 Argon cost?

Gemini 4 Argon's list price from Google DeepMind is $4 per million input tokens and $20 per million output tokens.

What is Gemini 4 Argon's context window?

Gemini 4 Argon supports a context window of up to 1M tokens.

How does Gemini 4 Argon perform on benchmarks?

On standardized evaluations tracked by Epoch AI, Gemini 4 Argon scores 82.4% on APEX, 96.4% on GPQA Diamond, 91.7% on SimpleBench.

Is Gemini 4 Argon open source?

Gemini 4 Argon is a proprietary model (API access), available through its provider's API rather than as a downloadable open-weight model.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Nov 19, 2026.