What is the IQ of DeepSeek V4.1 Flash?

DeepSeek V4.1 Flash has an estimated IQ of 126, which ranks #21 out of 173 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 DeepSeek V4.1 Flash best at?

Among the dimensions with published benchmark results, DeepSeek V4.1 Flash ranks highest in Abstract Reasoning (IQ 115, #19 of 197) and Computer Use (IQ 127, #21 of 173). Its lowest-ranked measured dimension is Reliability (IQ 118, #50 of 173).

What is the EQ of DeepSeek V4.1 Flash?

DeepSeek V4.1 Flash 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 DeepSeek V4.1 Flash. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.

Which benchmarks has DeepSeek V4.1 Flash been tested on?

DeepSeek V4.1 Flash has published results on 25 of the 39 benchmarks behind AI IQ's Composite IQ, including AA Long Context Reasoning v1.1, AutomationBench-AA, GDPval-AA v2.1 normalized score, Terminal-Bench 4.0 and Mystery Game Puzzles.

How much does DeepSeek V4.1 Flash cost?

DeepSeek V4.1 Flash costs $0.30 per million input tokens and $1.20 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $1.85 per million input-plus-output tokens.

Is DeepSeek V4.1 Flash open source?

Yes — DeepSeek V4.1 Flash is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.

Who developed DeepSeek V4.1 Flash?

DeepSeek V4.1 Flash was developed by DeepSeek, based in China.

When was DeepSeek V4.1 Flash released?

DeepSeek V4.1 Flash was publicly released on September 10, 2026.

How large is the context window of DeepSeek V4.1 Flash?

DeepSeek V4.1 Flash supports a context window of 1,000,000 tokens — the amount of text it can consider in a single request.

How fast is DeepSeek V4.1 Flash?

In independent measurements, DeepSeek V4.1 Flash generates about 209 tokens per second, with a typical response time of about 13 seconds.

DeepSeekReleased Sep 10, 2026

Open weights

DeepSeek V4.1 Flash

Strongest in Abstract Reasoning (#19 of 197) and Computer Use (#21 of 173); weakest in Reliability (#50 of 173). All seven dimensions are measured.

Composite IQ ?

126

#21 of 173 ranked models · ahead of 88%

DeepSeek's #2 model, behind DeepSeek V4.1 Flash 1017 (129)

Effective cost

$1.85/1M

$0.30 in · $1.20 out

Speed

209tok/s

13s typical response

Context

1Mtokens

1,000,000 tokens in

Evidence

7of 7 dims

25 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

115measured

#19 / 197

4/5

-12.5

Mathematical ReasoningMath

131measured

#30 / 173

4/5

-15

Academic ReasoningAcademic

130measured

#38 / 199

4/6

-5

Programmatic ReasoningCoding

135measured

#26 / 174

3/6

-13.5

Computer UseComputer Use

127measured

#21 / 173

1/5

-15

ReliabilityReliability

118measured

#50 / 173

4/6

-14.5

Knowledge WorkKnowledge Work

125measured

#30 / 173

5/6

-6

DeepSeek V4.1 Flash95% 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 DeepSeek V4.1 Flash. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 115 · measuredBest: #12 of 55 on Mystery Game Puzzles4/5

BenchmarkScorePercentileRankLeader · source

Mystery Game Puzzles

Full source leaderboard reconciliation vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 23

43.0%

#12 / 55

Leader gpt-6-astra · 84.0%
Full source leaderboard reconciliation vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 23

ARC-AGI-1

Full source leaderboard reconciliation · Oct 23

94.5%

#16 / 68

Leader fable-5 · 98.5%
Full source leaderboard reconciliation · Oct 23

ARC-AGI-2

Full source leaderboard reconciliation · Oct 23

72.9%

#19 / 68

Leader gpt-6-astra · 95.0%
Full source leaderboard reconciliation · Oct 23

ARC-AGI-3

Full source leaderboard reconciliation · Oct 23

0.9%

#12 / 23

Leader gpt-6-astra · 62.7%
Full source leaderboard reconciliation · Oct 23

Not yet reported

Chess Puzzles

Mathematical Reasoning IQ 131 · measuredBest: #17 of 61 on MathArena4/5

BenchmarkScorePercentileRankLeader · source

MathArena

Full source leaderboard reconciliation · Oct 23

42.3%

#17 / 61

Leader gpt-6.1-sol · 84.0%
Full source leaderboard reconciliation · Oct 23

FrontierMath Tier 1-3

Full source leaderboard reconciliation · Oct 23

67.4%

#28 / 83

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

FrontierMath Tier 4

Full source leaderboard reconciliation · Oct 23

26.8%

#32 / 71

Leader gpt-6.1-sol · 100%
Full source leaderboard reconciliation · Oct 23

ProofBench 1.1

Full source leaderboard reconciliation v1.1 · Oct 23

54.0%

#27 / 47

Leader fable-5.1 · 100%
Full source leaderboard reconciliation v1.1 · Oct 23

Not yet reported

AIME

Academic Reasoning IQ 130 · measuredBest: #45 of 157 on CritPt4/6

BenchmarkScorePercentileRankLeader · source

CritPt

Artificial Analysis v4.3.2 · Oct 23

14.3%

#45 / 157

Leader Muse Wonder Full · 32.9%
Artificial Analysis v4.3.2 · Oct 23

Humanity's Last Exam

Artificial Analysis v4.3.2 · Oct 23

39.2%

#51 / 165

Leader opus-5.5 · 61.4%
Artificial Analysis v4.3.2 · Oct 23

SciCode

Artificial Analysis v4.3.2 · Oct 23

51.9%

#52 / 165

Leader opus-5.5 · 66.9%
Artificial Analysis v4.3.2 · Oct 23

MMMU-Pro

Artificial Analysis v4.3.2 · Oct 23

77.0%

#41 / 87

Leader fable-5.1 · 90.6%
Artificial Analysis v4.3.2 · Oct 23

Not yet reported

GPQA DiamondMMLU-Pro

Programmatic Reasoning IQ 135 · measuredBest: #22 of 102 on Terminal-Bench 4.03/6

BenchmarkScorePercentileRankLeader · source

Terminal-Bench 4.0

Artificial Analysis v4.3.2 · Oct 23

26.8%

#22 / 102

Leader sonnet-5.5 · 63.6%
Artificial Analysis v4.3.2 · Oct 23

ProgramBench

Full source leaderboard reconciliation · Oct 23

16.5%

#17 / 61

Leader opus-5.5 · 65.0%
Full source leaderboard reconciliation · Oct 23

IOI V2

Full source leaderboard reconciliation v2 · Oct 23

40.3%

#33 / 42

Leader Gemini 4 Argon · 100%
Full source leaderboard reconciliation v2 · Oct 23

Not yet reported

LiveCodeBenchSWE-rebenchMirrorCode

Computer Use IQ 127 · measuredBest: #23 of 64 on Arena.ai Agent Arena1/5

BenchmarkScorePercentileRankLeader · source

Arena.ai Agent Arena

Arena.ai Agent Arena · Oct 23

2.6

#23 / 64

Leader opus-5.5 · 20.9
Arena.ai Agent Arena · Oct 23

Not yet reported

BrowseCompOSWorld-VerifiedMCP AtlasAgents' Last Exam

Reliability IQ 118 · measuredBest: #6 of 152 on AA Long Context Reasoning v1.14/6

BenchmarkScorePercentileRankLeader · source

AA Long Context Reasoning v1.1

Artificial Analysis v4.3.2 · Oct 23

84.0%

#6 / 152

Leader kimi-k3 · 88.7%
Artificial Analysis v4.3.2 · Oct 23

BullshitBench v2

Full source leaderboard reconciliation · Oct 23

57.0%

#24 / 83

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

AA Omniscience

Artificial Analysis v4.3.2 · Oct 23

-5.3

#78 / 156

Leader opus-5.5 · 46.4
Artificial Analysis v4.3.2 · Oct 23

SimpleQA Verified

Full source leaderboard reconciliation · Oct 23

42.9%

#43 / 79

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

Not yet reported

IFBenchFACTS Grounding

Knowledge Work IQ 125 · measuredBest: #6 of 104 on AutomationBench-AA5/6

BenchmarkScorePercentileRankLeader · source

AutomationBench-AA

Artificial Analysis v4.3.2 · Oct 23

68.9%

#6 / 104

Leader Gemini 4 Argon · 77.5%
Artificial Analysis v4.3.2 · Oct 23

GDPval-AA v2.1 normalized score

Artificial Analysis v4.3.2 · Oct 20

55.0%

#24 / 124

Leader opus-5.5 · 68.3%
Artificial Analysis v4.3.2 · Oct 20

AA-Briefcase v1.1 Elo

Artificial Analysis v4.3.2 · Oct 23

1,419

#28 / 109

Leader sonnet-5.5 · 1,816
Artificial Analysis v4.3.2 · Oct 23

GDP.pdf

Artificial Analysis v4.3.2 · Oct 23

12.8%

#45 / 104

Leader Muse Wonder Full · 32.6%
Artificial Analysis v4.3.2 · Oct 23

AA-AnalystAgent

Artificial Analysis vpass^5-2026-09-29 · Oct 18

35.0%

#23 / 30

Leader gemini-3.7-flash · 60.0%
Artificial Analysis vpass^5-2026-09-29 · Oct 18

Not yet reported

EnterpriseOps-Gym-AA

Release & availability

Facts from the publisher's announcement and documentation.

Released: Sep 10, 2026
Developer: DeepSeek
Access: The original September 10 MIT checkpoint remains downloadable. The mutable deepseek-flash hosted alias now serves DeepSeek V4.1 Flash 1017.
Weights: Open weightsDownloadable and self-hostable
Inputs: textimage
Context: 1,000,000 tokens
API price: $0.30 in · $1.20 outper 1M tokens · full pricing

What DeepSeek says

A new causal encoder-decoder mixture-of-experts architecture with native image and text input, text output, and a 1M-token context window.
The API supports thinking and non-thinking modes, tool calls, and Responses and Anthropic-compatible interfaces.

Limitations

Peak and off-peak API prices differ; the stored prices use peak rates.
The reported 552B parameter count describes the backbone; the model also has 196B of Engram conditional memory.
The original checkpoint is no longer served by the current deepseek-flash alias; retained hosted prices describe its historical peak rates.

Publisher statements, not AI IQ measurements.

Sourcesapi-docs.deepseek.com/news/news260910 api-docs.deepseek.com/quick_start/pricinghuggingface.co/deepseek-ai/DeepSeek-V4.1-Flashapi-docs.deepseek.com/updates

About this score

How much of DeepSeek V4.1 Flash's IQ is measured, and how the rest is estimated.

25/ 39benchmark results

7/ 7dimensions measured

0estimated

4/5Abstract

4/5Math

4/6Academic

3/6Coding

1/5Computer Use

4/6Reliability

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. Missing benchmarks can also draw on its predecessor, Deepseek V4 Flash; a predecessor is an estimation input, not evidence that this model has the predecessor's measured results.

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 DeepSeek V4.1 Flash

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

#1 overall

gpt-6.1-sol

IQ141 15 higherEffective cost$6.25 3.4× pricier

Predecessor

deepseek-v4-flash

IQ121 5 lowerEffective cost$1.68 1.1× cheaper

DeepSeek #1

DeepSeek V4.1 Flash 1017

IQ129 3 higherEffective cost$2.40 1.3× pricier

Closest IQ

Qwen3.8 Max 0902

IQ126 same IQEffective cost$20.42 11.0× pricier

Most similar profile

kimi-k3

IQ126 same IQEffective cost$20.02 10.8× pricier

Similar price

grok-4.3

IQ112 14 lowerEffective cost$1.68 1.1× cheaper

DeepSeek V4.1 FlashComparison model

Compare with any modelBrowse all comparisons →

Frequently asked questions

Quick answers about DeepSeek V4.1 Flash, 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 DeepSeek V4.1 Flash?: DeepSeek V4.1 Flash has an estimated IQ of 126, which ranks #21 out of 173 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 DeepSeek V4.1 Flash best at?: Among the dimensions with published benchmark results, DeepSeek V4.1 Flash ranks highest in Abstract Reasoning (IQ 115, #19 of 197) and Computer Use (IQ 127, #21 of 173). Its lowest-ranked measured dimension is Reliability (IQ 118, #50 of 173).
What is the EQ of DeepSeek V4.1 Flash?: DeepSeek V4.1 Flash 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 DeepSeek V4.1 Flash. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.
Which benchmarks has DeepSeek V4.1 Flash been tested on?: DeepSeek V4.1 Flash has published results on 25 of the 39 benchmarks behind AI IQ's Composite IQ, including AA Long Context Reasoning v1.1, AutomationBench-AA, GDPval-AA v2.1 normalized score, Terminal-Bench 4.0 and Mystery Game Puzzles.
How much does DeepSeek V4.1 Flash cost?: DeepSeek V4.1 Flash costs $0.30 per million input tokens and $1.20 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $1.85 per million input-plus-output tokens.
Is DeepSeek V4.1 Flash open source?: Yes — DeepSeek V4.1 Flash is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
Who developed DeepSeek V4.1 Flash?: DeepSeek V4.1 Flash was developed by DeepSeek, based in China.
When was DeepSeek V4.1 Flash released?: DeepSeek V4.1 Flash was publicly released on September 10, 2026.
How large is the context window of DeepSeek V4.1 Flash?: DeepSeek V4.1 Flash supports a context window of 1,000,000 tokens — the amount of text it can consider in a single request.
How fast is DeepSeek V4.1 Flash?: In independent measurements, DeepSeek V4.1 Flash generates about 209 tokens per second, with a typical response time of about 13 seconds.