Skip to content
Leaderboard / Google

Gemini 4 Argon

Google High CLOSED Proprietary
Verdict

Gemini 4 Argon is a top-tier model: #6 of 406 by SIscope score. Strong at everything it's been tested on, especially reasoning & knowledge. On the pricey side: costs more than 77% of priced models.

SIscope score
96
#6 of 406
Human-vote rank
#2
of 419 · score 1529
Price per 1M tokens, blended
$4
$2 in · $10 out
Context window
1M
1,048,576 tokens

Shape

Each spoke is a category, 0 at the centre, 100 at the edge.

CodingAgentsReasoningMathChatInstructions

Category scores

Percentiles: 90 means better than about 90% of tested models.

Coding96
Agents & tools89
Reasoning & knowledge100
Math96
Chat quality100
Following instructions100

What goes into each category · Compare with DeepSeek V4.1 Flash

Human-vote score over time

Text arena, recent updates.

152015301540Oct 15: 1532.3 (rank #1)Oct 20: 1532.3 (rank #1)Oct 22: 1525.7 (rank #1)Oct 28: 1528.6 (rank #2)Oct 15Oct 28

Coding votes over time

WebDev arena, recent updates.

166516751685Oct 16: 1678.2 (rank #9)Oct 18: 1678.1 (rank #9)Oct 19: 1678.3 (rank #10)Oct 22: 1672.7 (rank #11)Oct 23: 1674.2 (rank #11)Oct 27: 1679.9 (rank #10)Oct 28: 1679.2 (rank #11)Oct 16Oct 28

Independent test runs

Run by Epoch AI on the same harness for every model. Bars show the share of questions answered correctly.

GPQA Diamond96%
PhD-level science multiple choice questions
Better than 100% of 333 tested models. Run Oct 13, 2026.
SimpleQA Verified76%
Short factual questions; rewards not making things up
Better than 100% of 100 tested models. Run Oct 13, 2026.
OTIS Mock AIME 2024-25100%
Competition maths problems in the style of AIME
Better than 96% of 313 tested models. Run Oct 13, 2026.
FrontierMath Tiers 1-386%
Very hard, unpublished research-style maths problems
Better than 91% of 122 tested models. Run Oct 13, 2026.

Human-vote arenas

People compare two anonymous answers and pick the better one. Scores work like chess ratings.

ArenaRankScoreRange (95%)VotesUpdated
Text arena
Everyday chat, blind human votes
#2 of 41915291522–153511,889Oct 28, 2026
WebDev arena
Building web apps, blind human votes
#11 of 15216791668–16913,437Oct 28, 2026
Vision arena
Questions about images, blind human votes
Not on this board yet
Agent arena
Real multi-step agent sessions
#7 of 5710.6%9.8–11.5%52,038Nov 2, 2026

Source: Arena (formerly LMArena) leaderboard dataset, CC BY 4.0. Fetched Nov 4, 2026.

Price

Input
$2
per 1M tokens
Output
$10
per 1M tokens
Blended
$4
3 in : 1 out, per 1M
Context window
1M
max tokens per request
Speed
Coming soon
our own tests

List price from the OpenRouter models API (google/gemini-4-argon), Nov 4, 2026. Providers can charge differently.