Skip to content
Yardstick
  1. Models
  2. Google
  3. Gemini 4 Argon

Gemini 4 Argon

Google · Flagship · Sep 30, 2026NewPreview

Google's new frontier model, announced on 30 September 2026. So far it is limited to selected partners and is not yet on the public API.

Announcement (opens in a new tab)
Artificial Analysis Intelligence Index
52.6#5 of 29 · high effort
API price per 1M tokens
$2 / $10input / output · $4 blended
Context window
1Mtokens
Output speed
Not measured

Benchmark results

Each row shows this model (blue) against every other model in the catalogue (grey), worst on the left, best on the right.

Overall capability

  • Artificial Analysis Intelligence Index52.6#5 of 29

    High effort

    11.357.6

Coding

  • Arena WebDev1,678#6 of 27

    High effort, preliminary rating

    1,2631,814

Agents & professional work

  • GDPval-AA1,626#10 of 18

    High effort

    1,2421,866
  • Vending-Bench 2$13,718#2 of 9
    $2,158$15,515

Human preference

  • Arena Text1,525#1 of 24

    High effort, preliminary rating

    1,4251,525

Specs & pricing

Developer
Google
Released
Sep 30, 2026
Availability
Preview
Weights
Proprietary (API only)
Reasoning
Adjustable effort
Accepts
Text, Image, Video
Produces
Text
Context window
1,000,000 tokens
Input price
$2 / 1M tokens
Output price
$10 / 1M tokens

Introductory price announced by Google; it is due to rise to $4 input / $20 output per 1M tokens. The model is not yet purchasable through the public API.

Closest alternatives

Models from other labs nearest to Gemini 4 Argon on Artificial Analysis Intelligence Index.

  • GPT-6 Astra

    OpenAI

    Intelligence 52.7 · $10 / $50

    Compare side by side

  • GPT-6.1 Sol

    OpenAI

    Intelligence 51.8 · $2 / $10

    Compare side by side

  • Claude Fable 5.1

    Anthropic

    Intelligence 53.4 · $10 / $50

    Compare side by side

Sources

  • Announcement (opens in a new tab)
Yardstick

A plain reading of how today’s SI models compare. Scores are taken from each benchmark’s own leaderboard wherever one exists; anything a lab reported about its own model is marked.

Data as of Oct 8, 2026.

Explore

  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Methodology & sources

Sources

  • Artificial Analysis (opens in a new tab)
  • Arena (opens in a new tab)
  • Epoch AI Benchmarking Hub (opens in a new tab)
  • ARC Prize (opens in a new tab)
  • Terminal-Bench (opens in a new tab)
  • Scale AI Labs (opens in a new tab)

Yardstick is an independent reference and is not affiliated with, sponsored by or endorsed by any lab listed. Product names and logos are trademarks of their respective owners and are used only to identify them. Benchmarks are imperfect proxies: check a model on your own task before you commit to it.

  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs