GoogleReleased Sep 30, 2026
Proprietary

Gemini 4 Argon

Strongest in Academic Reasoning (#4 of 223) and Programmatic Reasoning (#5 of 187); weakest in Knowledge Work (#15 of 186). All seven dimensions are measured.

Composite IQ ?
139
#8 of 186 ranked models · ahead of 96%
Google's #2 model, behind Gemini 4 Carbon (142)
Effective cost
Not priced
No standard API rate
Speed
80tok/s
57s typical response
Context
Unverified
Not in publisher docs
Evidence
7of 7 dims
25 of 39 benchmarks

IQ profile

Seven dimensions, averaged equally into Composite IQ
Compare shape with
Measured95% range
#1 in dimensionTop-10 median

IQ dimensions

Ranks are among models scored in each dimension. Ranges are model-based estimates, not calibrated confidence intervals.

DimensionIQPosition · 70–160RankEvidencevs Top 10
132measured
#6 / 221
3/5
-0.5
144measured
#11 / 186
4/5
-5.5
136measured
#4 / 223
3/6
+1
152measured
#5 / 187
3/6
+1
144measured
#10 / 186
2/5
-2.5
134measured
#9 / 188
5/6
-1.5
131measured
#15 / 186
5/6
-1
Gemini 4 Argon95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

25 of 39 Composite IQ benchmarks have a published result for Gemini 4 Argon. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 132 · measuredBest: #4 of 73 on ARC-AGI-13/5
BenchmarkScorePercentileRankLeader · source
ARC-AGI-1
Full source leaderboard reconciliation · Oct 23
98.5%
#4 / 73
Leader fable-5 · 98.5%
Full source leaderboard reconciliation · Oct 23
ARC-AGI-2
Full source leaderboard reconciliation · Oct 23
91.3%
#8 / 73
Leader gpt-6.2-astra · 95.8%
Full source leaderboard reconciliation · Oct 23
ARC-AGI-3
Full source leaderboard reconciliation · Oct 23
49.8%
#7 / 29
Leader fable-5.2 · 81.2%
Full source leaderboard reconciliation · Oct 23
Not yet reported
Chess PuzzlesMystery Game Puzzles
Mathematical Reasoning IQ 144 · measuredBest: #5 of 49 on ProofBench 1.14/5
BenchmarkScorePercentileRankLeader · source
ProofBench 1.1
Full source leaderboard reconciliation v1.1 · Oct 27
99.0%
#5 / 49
Leader fable-5.1 · 100%
Full source leaderboard reconciliation v1.1 · Oct 27
MathArena
Full source leaderboard reconciliation · Nov 13
62.1%
#11 / 70
Leader gpt-6.2-astra · 92.2%
Full source leaderboard reconciliation · Nov 13
FrontierMath Tier 1-3
Full source leaderboard reconciliation · Oct 23
86.0%
#14 / 89
Leader fable-5.2 · 95.1%
Full source leaderboard reconciliation · Oct 23
FrontierMath Tier 4
Full source leaderboard reconciliation · Oct 23
78.0%
#14 / 77
Leader fable-5.2 · 100%
Full source leaderboard reconciliation · Oct 23
Not yet reported
AIME
Academic Reasoning IQ 136 · measuredBest: #3 of 167 on Humanity's Last Exam3/6
BenchmarkScorePercentileRankLeader · source
Humanity's Last Exam
Artificial Analysis v4.3.2 · Oct 27
57.1%
#3 / 167
Leader opus-5.5 · 61.4%
Artificial Analysis v4.3.2 · Oct 27
SciCode
Artificial Analysis v4.3.2 · Oct 27
61.8%
#4 / 167
Leader opus-5.5 · 66.9%
Artificial Analysis v4.3.2 · Oct 27
CritPt
Artificial Analysis v4.3.2 · Oct 27
27.1%
#14 / 159
Leader Muse Wonder Full · 32.9%
Artificial Analysis v4.3.2 · Oct 27
Not yet reported
GPQA DiamondMMLU-ProMMMU-Pro
Programmatic Reasoning IQ 152 · measuredBest: #1 of 43 on IOI V23/6
BenchmarkScorePercentileRankLeader · source
IOI V2
Full source leaderboard reconciliation v2 · Oct 27
100%
#1 / 43
Leader on this benchmark
Full source leaderboard reconciliation v2 · Oct 27
Terminal-Bench 4.0
Artificial Analysis v5.0.1 · Nov 13
57.1%
#7 / 109
Leader Kimi K3.1 Pro · 67.2%
Artificial Analysis v5.0.1 · Nov 13
ProgramBench
Full source leaderboard reconciliation · Oct 23
41.0%
#8 / 63
Leader opus-5.5 · 65.0%
Full source leaderboard reconciliation · Oct 23
Not yet reported
LiveCodeBenchSWE-rebenchMirrorCode
Computer Use IQ 144 · measuredBest: #1 of 35 on MCP Atlas2/5
BenchmarkScorePercentileRankLeader · source
MCP Atlas
Full source leaderboard reconciliation · Nov 6
90.7%
#1 / 35
Leader on this benchmark
Full source leaderboard reconciliation · Nov 6
Arena.ai Agent Arena
Arena.ai Agent Arena · Nov 19
10.3
#13 / 77
Leader fable-5.2 · 27.4
Arena.ai Agent Arena · Nov 19
Not yet reported
BrowseCompOSWorld-VerifiedAgents' Last Exam
Reliability IQ 134 · measuredBest: #5 of 158 on AA Omniscience5/6
BenchmarkScorePercentileRankLeader · source
AA Omniscience
Artificial Analysis v4.3.2 · Oct 27
42.4
#5 / 158
Leader opus-5.5 · 46.4
Artificial Analysis v4.3.2 · Oct 27
SimpleQA Verified
Full source leaderboard reconciliation · Oct 23
76.0%
#4 / 85
Leader fable-5.2 · 94.0%
Full source leaderboard reconciliation · Oct 23
BullshitBench v2
Full source leaderboard reconciliation · Oct 23
65.0%
#20 / 96
Leader opus-4.8 · 94.0%
Full source leaderboard reconciliation · Oct 23
AA Long Context Reasoning v1.1
Artificial Analysis v4.3.2 · Oct 27
79.7%
#50 / 154
Leader kimi-k3 · 88.7%
Artificial Analysis v4.3.2 · Oct 27
FACTS Grounding
Full source leaderboard reconciliation · Nov 13
73.4%
#16 / 47
Leader gpt-6.1-terra · 84.2%
Full source leaderboard reconciliation · Nov 13
Not yet reported
IFBench
Knowledge Work IQ 131 · measuredBest: #2 of 112 on AutomationBench-AA5/6
BenchmarkScorePercentileRankLeader · source
AutomationBench-AA
Artificial Analysis v5.0.1 · Nov 13
77.5%
#2 / 112
Leader Gemini 4 Carbon · 77.8%
Artificial Analysis v5.0.1 · Nov 13
GDPval-AA v2.1 normalized score
Artificial Analysis v4.3.2 · Oct 23
56.2%
#15 / 124
Leader opus-5.5 · 68.3%
Artificial Analysis v4.3.2 · Oct 23
GDP.pdf
Artificial Analysis v5.0.1 · Nov 13
21.8%
#24 / 111
Leader Muse Wonder Full · 32.6%
Artificial Analysis v5.0.1 · Nov 13
AA-Briefcase v1.1 Elo
Artificial Analysis v5.0.1 · Nov 13
1,484
#26 / 116
Leader sonnet-5.5 · 1,816
Artificial Analysis v5.0.1 · Nov 13
AA-AnalystAgent
Full source leaderboard reconciliation vpass^5-2026-09-29 · Oct 27
48.8%
#12 / 35
Leader gemini-3.7-flash · 60.0%
Full source leaderboard reconciliation vpass^5-2026-09-29 · Oct 27
Not yet reported
EnterpriseOps-Gym-AA

Release & availability

Facts from the publisher's announcement and documentation.

Released
Developer
Google
Access
Generally available through the paid Gemini API and Google AI Studio as gemini-4-argon since October 13, 2026.
Weights
ProprietaryHosted access only
Inputs
textimagevideoaudio
Context
Not published1,048,576 tokens shared by input and output; the remaining input budget depends on output length.
API price
No per-token rateCurrent introductory paid Standard rates through November 30, 2026 are $2 input and $10 output per million tokens, with $0.10 implicit cached input. Announced rates from December 1 are $4/$20 and $0.20 cached input; introductory and future rates are kept out of stored numeric standard pricing.

What Google says

  • Google first announced Gemini 4 Argon on September 30, 2026 and made it generally available on October 13.
  • Supports text, image, video, audio and PDF input with text output.
  • The documented 1,048,576-token context budget is shared by input and output; long decode can use up to that budget.

Limitations

  • Not available on the Gemini API Free tier or the Google AI Pro plan.
  • The context limit is a unified input and output budget, not a separate one-million-token input allowance.
  • Introductory rates end November 30; future December rates are not current standard prices.
  • Provider harness results do not substitute for independent measurements.

Publisher statements, not AI IQ measurements.

About this score

How much of Gemini 4 Argon's IQ is measured, and how the rest is estimated.

25/ 39benchmark results
7/ 7dimensions measured
0estimated
3/5Abstract
4/5Math
3/6Academic
3/6Coding
2/5Computer Use
5/6Reliability
5/6Knowledge Work
Bar height = share of benchmarks with resultsEstimatedNo score
Measured results are shown as published

Every score in the benchmark tables comes from its source unchanged. Composite IQ averages all seven dimensions equally.

How gaps are handled

Every dimension has at least one published result.

A missing result is not a zero

Benchmark versions and evaluation harnesses can differ between sources, so small gaps between models may not be meaningful.

Read the scoring methodology →

Compare Gemini 4 Argon

Side-by-side pages for the models it is most often weighed against.

Gemini 4 ArgonComparison model
Browse all comparisons →

Frequently asked questions

Quick answers about Gemini 4 Argon, drawn from the same data as the rest of this page.

How AI IQ scores models →Get this data from the API →
What is the IQ of Gemini 4 Argon?
Gemini 4 Argon has an estimated IQ of 139, which ranks #8 out of 186 public models on the AI IQ leaderboard. The score combines seven equally weighted reasoning dimensions using measured benchmark results and conservative estimates for missing coverage.
What is Gemini 4 Argon best at?
Among the dimensions with published benchmark results, Gemini 4 Argon ranks highest in Academic Reasoning (IQ 136, #4 of 223) and Programmatic Reasoning (IQ 152, #5 of 187). Its lowest-ranked measured dimension is Knowledge Work (IQ 131, #15 of 186).
What is the EQ of Gemini 4 Argon?
Gemini 4 Argon doesn't have an EQ score yet. EQ (emotional reasoning) is calculated from EQ-Bench 4 and AttuneBench, and neither has published a result for Gemini 4 Argon. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.
Which benchmarks has Gemini 4 Argon been tested on?
Gemini 4 Argon has published results on 25 of the 39 benchmarks behind AI IQ's Composite IQ, including AutomationBench-AA, Humanity's Last Exam, IOI V2, SciCode and MCP Atlas.
How much does Gemini 4 Argon cost?
AI IQ doesn't have a verified standard per-token API price for Gemini 4 Argon, so it can't calculate an effective cost. This is common for models offered only through subscriptions, restricted-access programs, or self-hosting.
Is Gemini 4 Argon open source?
No — Gemini 4 Argon is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.
Who developed Gemini 4 Argon?
Gemini 4 Argon was developed by Google, based in the United States.
When was Gemini 4 Argon released?
Gemini 4 Argon was publicly released on September 30, 2026.
How large is the context window of Gemini 4 Argon?
AI IQ hasn't verified an input context limit for Gemini 4 Argon from its developer's documentation, so none is listed rather than guessing.
How fast is Gemini 4 Argon?
In independent measurements, Gemini 4 Argon generates about 80 tokens per second, with a typical response time of about 57 seconds.