What is the IQ of Solar Mini 4?

Solar Mini 4 has an estimated IQ of 107, which ranks #92 out of 176 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 Solar Mini 4 best at?

Among the dimensions with published benchmark results, Solar Mini 4 ranks highest in Reliability (IQ 116, #67 of 177) and Knowledge Work (IQ 112, #80 of 176). Its lowest-ranked measured dimension is Programmatic Reasoning (IQ 109, #100 of 177).

What is the EQ of Solar Mini 4?

Solar Mini 4 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 Solar Mini 4. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.

Which benchmarks has Solar Mini 4 been tested on?

Solar Mini 4 has published results on 10 of the 39 benchmarks behind AI IQ's Composite IQ, including AA Long Context Reasoning v1.1, SciCode, GDPval-AA v2.1 normalized score, AutomationBench-AA and AA-Briefcase v1.1 Elo.

How much does Solar Mini 4 cost?

Solar Mini 4 costs $0.10 per million input tokens and $0.40 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $1.07 per million input-plus-output tokens.

Is Solar Mini 4 open source?

No — Solar Mini 4 is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.

Who developed Solar Mini 4?

Solar Mini 4 was developed by Upstage, based in South Korea.

When was Solar Mini 4 released?

Solar Mini 4 was publicly released on September 22, 2026.

How large is the context window of Solar Mini 4?

Solar Mini 4 supports a context window of 524,288 tokens — the amount of text it can consider in a single request.

How fast is Solar Mini 4?

In independent measurements, Solar Mini 4 generates about 122 tokens per second, with a typical response time of about 22 seconds.

UpstageReleased Sep 22, 2026

Proprietary

Solar Mini 4

Its best results are in Reliability (#67 of 177) and Knowledge Work (#80 of 176); weakest in Programmatic Reasoning (#100 of 177). Three of seven dimensions are estimated, because no results are published yet.

Composite IQ ?

107

#92 of 176 ranked models · ahead of 48%

World (Other)'s #7 model, behind mistral-large-4 (119)

Effective cost

$1.07/1M

$0.10 in · $0.40 out

Speed

122tok/s

22s typical response

Context

524Ktokens

524,288 tokens in

Evidence

4of 7 dims

10 of 39 benchmarks

Compare with gpt-6.1-solAPI pricing

IQ profile

Seven dimensions, averaged equally into Composite IQ

Compare shape with

MeasuredEstimated95% 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

Abstract ReasoningAbstract

94estimated

#85 / 201

0/5

-32.5

Mathematical ReasoningMath

97estimated

#131 / 176

0/5

-50

Academic ReasoningAcademic

118measured

#114 / 203

3/6

-17.5

Programmatic ReasoningCoding

109measured

#100 / 177

1/6

-40.5

Computer UseComputer Use

104estimated

#106 / 176

0/5

-39

ReliabilityReliability

116measured

#67 / 177

2/6

-18

Knowledge WorkKnowledge Work

112measured

#80 / 176

4/6

-20

Solar Mini 495% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

10 of 39 Composite IQ benchmarks have a published result for Solar Mini 4. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 94 · estimatedNo published results0/5

No published results yet for Abstract Reasoning. The 94 shown is estimated from Solar Mini 4's other dimensions and will be replaced as results come in. How estimates work

ARC-AGI-3ARC-AGI-2ARC-AGI-1Chess PuzzlesMystery Game Puzzles

Mathematical Reasoning IQ 97 · estimatedNo published results0/5

No published results yet for Mathematical Reasoning. The 97 shown is estimated from Solar Mini 4's other dimensions and will be replaced as results come in. How estimates work

FrontierMath Tier 4FrontierMath Tier 1-3ProofBench 1.1MathArenaAIME

Academic Reasoning IQ 118 · measuredBest: #78 of 167 on SciCode3/6

BenchmarkScorePercentileRankLeader · source

SciCode

Artificial Analysis v4.3.2 · Oct 27

47.6%

#78 / 167

Leader opus-5.5 · 66.9%
Artificial Analysis v4.3.2 · Oct 27

Humanity's Last Exam

Artificial Analysis v4.3.2 · Oct 27

25.8%

#103 / 167

Leader opus-5.5 · 61.4%
Artificial Analysis v4.3.2 · Oct 27

CritPt

Artificial Analysis v4.3.2 · Oct 27

1.4%

#111 / 159

Leader Muse Wonder Full · 32.9%
Artificial Analysis v4.3.2 · Oct 27

Not yet reported

GPQA DiamondMMLU-ProMMMU-Pro

Programmatic Reasoning IQ 109 · measuredBest: #63 of 106 on Terminal-Bench 4.01/6

BenchmarkScorePercentileRankLeader · source

Terminal-Bench 4.0

Artificial Analysis v5.0.1 · Nov 11

1.0%

#63 / 106

Leader Kimi K3.1 Pro · 67.2%
Artificial Analysis v5.0.1 · Nov 11

Not yet reported

LiveCodeBenchIOI V2ProgramBenchSWE-rebenchMirrorCode

Computer Use IQ 104 · estimatedNo published results0/5

No published results yet for Computer Use. The 104 shown is estimated from Solar Mini 4's other dimensions and will be replaced as results come in. How estimates work

BrowseCompOSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam

Reliability IQ 116 · measuredBest: #15 of 154 on AA Long Context Reasoning v1.12/6

BenchmarkScorePercentileRankLeader · source

AA Long Context Reasoning v1.1

Artificial Analysis v4.3.2 · Oct 27

83.3%

#15 / 154

Leader kimi-k3 · 88.7%
Artificial Analysis v4.3.2 · Oct 27

AA Omniscience

Artificial Analysis v4.3.2 · Oct 27

-10.8

#94 / 158

Leader opus-5.5 · 46.4
Artificial Analysis v4.3.2 · Oct 27

Not yet reported

SimpleQA VerifiedBullshitBench v2IFBenchFACTS Grounding

Knowledge Work IQ 112 · measuredBest: #63 of 124 on GDPval-AA v2.1 normalized score4/6

BenchmarkScorePercentileRankLeader · source

GDPval-AA v2.1 normalized score

Artificial Analysis v4.3.2 · Oct 23

29.1%

#63 / 124

Leader opus-5.5 · 68.3%
Artificial Analysis v4.3.2 · Oct 23

AutomationBench-AA

Artificial Analysis v5.0.1 · Nov 11

22.3%

#57 / 109

Leader Gemini 4 Carbon · 77.8%
Artificial Analysis v5.0.1 · Nov 11

AA-Briefcase v1.1 Elo

Artificial Analysis v5.0.1 · Nov 11

868

#67 / 113

Leader sonnet-5.5 · 1,816
Artificial Analysis v5.0.1 · Nov 11

GDP.pdf

Artificial Analysis v5.0.1 · Nov 11

5.0%

#84 / 108

Leader Muse Wonder Full · 32.6%
Artificial Analysis v5.0.1 · Nov 11

Not yet reported

AA-AnalystAgentEnterpriseOps-Gym-AA

Release & availability

Facts from the publisher's announcement and documentation.

Released: Sep 22, 2026
Developer: Upstage
Access: Available through Upstage Console and API as solar-mini4 (solar-mini4-260922), with a public Playground.
Weights: ProprietaryHosted access only
Inputs: text
Context: 524,288 tokens
API price: $0.10 in · $0.40 outper 1M tokens · full pricing

What Upstage says

Compact mixture-of-experts model with 35B total and 3B active parameters.
Supports reasoning, structured outputs, tool calling, and English, Korean, and Japanese.
Supports 512K context and up to 128K output tokens.

Limitations

Text-only model; no published open weights.
Canonical prices are standard rates; a temporary 50% discount lasts through October 22, 2026 UTC.

Publisher statements, not AI IQ measurements.

Sourcesconsole.upstage.ai/docs/models/solar-mini-4 console.upstage.ai/docs/resources/changelogupstage.ai/pricing

About this score

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

10/ 39benchmark results

4/ 7dimensions measured

3estimated

0/5Abstract

0/5Math

3/6Academic

1/6Coding

0/5Computer Use

2/6Reliability

4/6Knowledge 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.

Estimates fill the gaps

Abstract Reasoning, Mathematical Reasoning and Computer Use have no published results yet, so their IQ is estimated from Solar Mini 4's other dimensions.

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 Solar Mini 4

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

#1 overall

gpt-6.1-sol

IQ141 34 higherEffective cost$6.30 5.9× pricier

World (Other) #1

mistral-large-4

IQ119 12 higherEffective cost$4.69 4.4× pricier

Closest IQ

inkling

IQ107 same IQEffective cost$4.79 4.5× pricier

Most similar profile

nemotron-3-ultra

IQ105 2 lowerEffective cost$2.12 2.0× pricier

Similar price

trinity-large

IQ93 14 lowerEffective cost$1.04 about the same

#2 overall

opus-5.5

IQ141 34 higherEffective cost$35.73 33.3× pricier

Solar Mini 4Comparison model

Compare with any modelBrowse all comparisons →

Frequently asked questions

Quick answers about Solar Mini 4, 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 Solar Mini 4?: Solar Mini 4 has an estimated IQ of 107, which ranks #92 out of 176 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 Solar Mini 4 best at?: Among the dimensions with published benchmark results, Solar Mini 4 ranks highest in Reliability (IQ 116, #67 of 177) and Knowledge Work (IQ 112, #80 of 176). Its lowest-ranked measured dimension is Programmatic Reasoning (IQ 109, #100 of 177).
What is the EQ of Solar Mini 4?: Solar Mini 4 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 Solar Mini 4. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.
Which benchmarks has Solar Mini 4 been tested on?: Solar Mini 4 has published results on 10 of the 39 benchmarks behind AI IQ's Composite IQ, including AA Long Context Reasoning v1.1, SciCode, GDPval-AA v2.1 normalized score, AutomationBench-AA and AA-Briefcase v1.1 Elo.
How much does Solar Mini 4 cost?: Solar Mini 4 costs $0.10 per million input tokens and $0.40 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $1.07 per million input-plus-output tokens.
Is Solar Mini 4 open source?: No — Solar Mini 4 is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.
Who developed Solar Mini 4?: Solar Mini 4 was developed by Upstage, based in South Korea.
When was Solar Mini 4 released?: Solar Mini 4 was publicly released on September 22, 2026.
How large is the context window of Solar Mini 4?: Solar Mini 4 supports a context window of 524,288 tokens — the amount of text it can consider in a single request.
How fast is Solar Mini 4?: In independent measurements, Solar Mini 4 generates about 122 tokens per second, with a typical response time of about 22 seconds.