What is the IQ of qwen3.5-397b?

qwen3.5-397b has an estimated IQ of 109, which ranks #87 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 qwen3.5-397b best at?

Among the dimensions with published benchmark results, qwen3.5-397b ranks highest in Reliability (IQ 117, #62 of 177) and Computer Use (IQ 114, #71 of 176). Its lowest-ranked measured dimension is Abstract Reasoning (IQ 81, #152 of 201).

What is the EQ of qwen3.5-397b?

qwen3.5-397b has an estimated EQ of 108. EQ measures emotional reasoning — how well a model reads and responds to emotional context — and is scored separately from IQ.

Which benchmarks has qwen3.5-397b been tested on?

qwen3.5-397b has published results on 20 of the 39 benchmarks behind AI IQ's Composite IQ, including BullshitBench v2, IFBench, LiveCodeBench, GPQA Diamond and MMMU-Pro.

How much does qwen3.5-397b cost?

qwen3.5-397b costs $0.60 per million input tokens and $3.60 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $4.90 per million input-plus-output tokens.

Is qwen3.5-397b open source?

Yes — qwen3.5-397b is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.

Who developed qwen3.5-397b?

qwen3.5-397b was developed by Alibaba, based in China.

When was qwen3.5-397b released?

qwen3.5-397b was publicly released on February 16, 2026.

How large is the context window of qwen3.5-397b?

qwen3.5-397b supports a context window of 262,000 tokens — the amount of text it can consider in a single request.

How fast is qwen3.5-397b?

In independent measurements, qwen3.5-397b generates about 58 tokens per second, with a typical response time of about 65 seconds.

AlibabaReleased Feb 16, 2026

Open weights

qwen3.5-397b

Its best results are in Reliability (#62 of 177) and Computer Use (#71 of 176); weakest in Abstract Reasoning (#152 of 201). All seven dimensions are measured.

Composite IQ ?

109

#87 of 176 ranked models · ahead of 51%

Alibaba's #11 model, behind Qwen3.8 Max 0902 (125)

Effective cost

$4.90/1M

$0.60 in · $3.60 out

Speed

58tok/s

65s typical response

Context

262Ktokens

262,000 tokens in

Evidence

7of 7 dims

20 of 39 benchmarks

Compare with gpt-6.1-solAPI pricing

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

Abstract ReasoningAbstract

81measured

#152 / 201

1/5

-45.5

Mathematical ReasoningMath

113measured

#81 / 176

2/5

-34

Academic ReasoningAcademic

121measured

#97 / 203

5/6

-14.5

Programmatic ReasoningCoding

110measured

#97 / 177

2/6

-39.5

Computer UseComputer Use

114measured

#71 / 176

1/5

-29

ReliabilityReliability

117measured

#62 / 177

5/6

-17

Knowledge WorkKnowledge Work

109measured

#90 / 176

4/6

-22.5

qwen3.5-397b95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

20 of 39 Composite IQ benchmarks have a published result for qwen3.5-397b. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 81 · measuredBest: #69 of 76 on Chess Puzzles1/5

BenchmarkScorePercentileRankLeader · source

Chess Puzzles

Epoch AI vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 15

12.0%

#69 / 76

Leader gpt-6-astra · 72.0%
Epoch AI vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 15

Not yet reported

ARC-AGI-3ARC-AGI-2ARC-AGI-1Mystery Game Puzzles

Mathematical Reasoning IQ 113 · measuredBest: #43 of 64 on MathArena2/5

BenchmarkScorePercentileRankLeader · source

MathArena

Full source leaderboard reconciliation · Oct 28

18.8%

#43 / 64

Leader gpt-6.1-sol · 84.0%
Full source leaderboard reconciliation · Oct 28

FrontierMath Tier 1-3

Full source leaderboard reconciliation · Oct 23

31.2%

#64 / 84

Leader gpt-6-astra · 93.7%
Full source leaderboard reconciliation · Oct 23

Not yet reported

FrontierMath Tier 4ProofBench 1.1AIME

Academic Reasoning IQ 121 · measuredBest: #54 of 147 on GPQA Diamond5/6

BenchmarkScorePercentileRankLeader · source

GPQA Diamond

Artificial Analysis v4.3.2 · Oct 27

89.3%

#54 / 147

Leader gpt-6-astra · 96.1%
Artificial Analysis v4.3.2 · Oct 27

MMMU-Pro

Artificial Analysis v4.3.2 · Oct 27

77.3%

#41 / 89

Leader fable-5.1 · 90.6%
Artificial Analysis v4.3.2 · Oct 27

Humanity's Last Exam

Artificial Analysis v4.3.2 · Oct 27

29.0%

#91 / 167

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

SciCode

Artificial Analysis v4.3.2 · Oct 27

44.8%

#94 / 167

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

CritPt

Artificial Analysis v4.3.2 · Oct 27

1.7%

#108 / 159

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

Not yet reported

MMLU-Pro

Programmatic Reasoning IQ 110 · measuredBest: #15 of 61 on LiveCodeBench2/6

BenchmarkScorePercentileRankLeader · source

LiveCodeBench

Vals.ai · Aug 4

85.3%

#15 / 61

Leader Fugu Ultra v1.0 · 93.2%
Vals.ai · Aug 4

Terminal-Bench 4.0

Artificial Analysis v4.3.2 · Oct 27

0.0%

#100 / 104

Leader Kimi K3.1 Pro · 67.2%
Artificial Analysis v4.3.2 · Oct 27

Not yet reported

IOI V2ProgramBenchSWE-rebenchMirrorCode

Computer Use IQ 114 · measuredBest: #30 of 43 on BrowseComp1/5

BenchmarkScorePercentileRankLeader · source

BrowseComp

Full source leaderboard reconciliation · Oct 23

69.0%

#30 / 43

Leader Kimi K3.1 Pro · 92.1%
Full source leaderboard reconciliation · Oct 23

Not yet reported

OSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam

Reliability IQ 117 · measuredBest: #5 of 83 on BullshitBench v25/6

BenchmarkScorePercentileRankLeader · source

BullshitBench v2

Full source leaderboard reconciliation · Oct 23

78.0%

#5 / 83

Leader opus-4.8 · 94.0%
Full source leaderboard reconciliation · Oct 23

IFBench

Artificial Analysis v4.3.2 · Oct 27

78.8%

#9 / 96

Leader minimax-m3 · 82.9%
Artificial Analysis v4.3.2 · Oct 27

AA Long Context Reasoning v1.1

Artificial Analysis v4.3.2 · Oct 27

77.3%

#77 / 154

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

AA Omniscience

Artificial Analysis v4.3.2 · Oct 27

-30.7

#125 / 158

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

SimpleQA Verified

Full source leaderboard reconciliation · Oct 23

25.4%

#69 / 80

Leader Gemini 4 Argon · 76.0%
Full source leaderboard reconciliation · Oct 23

Not yet reported

FACTS Grounding

Knowledge Work IQ 109 · measuredBest: #50 of 106 on GDP.pdf4/6

BenchmarkScorePercentileRankLeader · source

GDP.pdf

Artificial Analysis v4.3.2 · Oct 27

12.2%

#50 / 106

Leader Muse Wonder Full · 32.6%
Artificial Analysis v4.3.2 · Oct 27

AutomationBench-AA

Artificial Analysis v4.3.2 · Oct 27

6.9%

#70 / 106

Leader Gemini 4 Argon · 77.5%
Artificial Analysis v4.3.2 · Oct 27

GDPval-AA v2.1 normalized score

Artificial Analysis v4.3.2 · Oct 23

14.5%

#88 / 124

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

AA-Briefcase v1.1 Elo

Artificial Analysis v4.3.2 · Oct 27

562

#86 / 111

Leader sonnet-5.5 · 1,812
Artificial Analysis v4.3.2 · Oct 27

Not yet reported

AA-AnalystAgentEnterpriseOps-Gym-AA

Release & availability

Release details on record for this model.

Released: Feb 16, 2026
Developer: Alibaba
Weights: Open weightsDownloadable and self-hostable
Context: 262,000 tokens
API price: $0.60 in · $3.60 outper 1M tokens · full pricing

About this score

How much of qwen3.5-397b's IQ is measured, and how the rest is estimated.

20/ 39benchmark results

7/ 7dimensions measured

0estimated

1/5Abstract

2/5Math

5/6Academic

2/6Coding

1/5Computer Use

5/6Reliability

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.

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 qwen3.5-397b

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

#1 overall

gpt-6.1-sol

IQ141 32 higherEffective cost$6.30 1.3× pricier

Alibaba #1

Qwen3.8 Max 0902

IQ125 16 higherEffective cost$20.79 4.2× pricier

Closest IQ

Fugu Ultra v2.0

IQ109 same IQEffective cost$44.78 9.1× pricier

Most similar profile

kimi-k2.5

IQ110 1 higherEffective cost$2.79 1.8× cheaper

Similar price

inkling

IQ107 2 lowerEffective cost$4.79 about the same

#2 overall

opus-5.5

IQ141 32 higherEffective cost$35.73 7.3× pricier

qwen3.5-397bComparison model

Compare with any modelBrowse all comparisons →

Frequently asked questions

Quick answers about qwen3.5-397b, 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 qwen3.5-397b?: qwen3.5-397b has an estimated IQ of 109, which ranks #87 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 qwen3.5-397b best at?: Among the dimensions with published benchmark results, qwen3.5-397b ranks highest in Reliability (IQ 117, #62 of 177) and Computer Use (IQ 114, #71 of 176). Its lowest-ranked measured dimension is Abstract Reasoning (IQ 81, #152 of 201).
What is the EQ of qwen3.5-397b?: qwen3.5-397b has an estimated EQ of 108. EQ measures emotional reasoning — how well a model reads and responds to emotional context — and is scored separately from IQ.
Which benchmarks has qwen3.5-397b been tested on?: qwen3.5-397b has published results on 20 of the 39 benchmarks behind AI IQ's Composite IQ, including BullshitBench v2, IFBench, LiveCodeBench, GPQA Diamond and MMMU-Pro.
How much does qwen3.5-397b cost?: qwen3.5-397b costs $0.60 per million input tokens and $3.60 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $4.90 per million input-plus-output tokens.
Is qwen3.5-397b open source?: Yes — qwen3.5-397b is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
Who developed qwen3.5-397b?: qwen3.5-397b was developed by Alibaba, based in China.
When was qwen3.5-397b released?: qwen3.5-397b was publicly released on February 16, 2026.
How large is the context window of qwen3.5-397b?: qwen3.5-397b supports a context window of 262,000 tokens — the amount of text it can consider in a single request.
How fast is qwen3.5-397b?: In independent measurements, qwen3.5-397b generates about 58 tokens per second, with a typical response time of about 65 seconds.