Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models799Deals30d267Policy30d81, 17 pendingSources1,138

DeepSeek-V4.1-Flash

DeepSeekReleased Sep 10, 2026Current

Score basis

DeepSeek-V4.1-Flash is strong in long context; capable in reasoning and coding; and behind the leaders in agentic tasks, multimodal tasks, math, factuality, and instruction following. Too few results yet to rate safety or multilingual tasks.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Strong1 capability
  1. Long Context−4.9%1 of 3
Capable2 capabilities
  1. Reasoning−14.0%6 of 6
  2. Coding−19.2%5 of 10
Limited5 capabilities
  1. Agentic−29.7%3 of 7
  2. Multimodal−31.2%2 of 6
  3. Math−32.5%5 of 5
  4. Factuality−38.6%3 of 4
  5. Instruction Following−39.0%1 of 3
Not rated2 capabilities

Too few results yet to rate Safety or Multilingual.

Price

$0.30input$1.20outputper million tokens

From Artificial Analysis · 5 providers tracked · All prices

Cheaper than 56% of 322 priced models · 3:1 input-to-output blend, log scale

Evidence

110results on53benchmarks

  • 49 independently verified
  • 23 aggregator
  • 26 vendor-reported
  • 12 cross-referenced

From 18 sources · latest Oct 27, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Research

17 papers reference DeepSeek-V4.1-Flash

DeepSeek-V4.1-Flash benchmark results

110 results on 53 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Long Context

StrongFull long context ranking

4.9% behind the leader1 of 3 ranked benchmarks measured

Show 1 more long context resultHide 1 long context result

Reasoning

CapableFull reasoning ranking

14.0% behind the leader6 of 6 ranked benchmarks measured

Show 20 more reasoning resultsHide 20 reasoning results

Coding

CapableFull coding ranking

19.2% behind the leader5 of 10 ranked benchmarks measured

Show 3 more coding resultsHide 3 coding results

Agentic

LimitedFull agentic ranking

29.7% behind the leader3 of 7 ranked benchmarks measured

Show 3 more agentic resultsHide 3 agentic results

Multimodal

LimitedFull multimodal ranking

31.2% behind the leader2 of 6 ranked benchmarks measured

Show 3 more multimodal resultsHide 3 multimodal results

Math

LimitedFull math ranking

32.5% behind the leader5 of 5 ranked benchmarks measured

Show 2 more math resultsHide 2 math results

Factuality

LimitedFull factuality ranking

38.6% behind the leader3 of 4 ranked benchmarks measured

Show 4 more factuality resultsHide 4 factuality results

Instruction Following

LimitedFull instruction following ranking

39.0% behind the leader1 of 3 ranked benchmarks measured

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 45 more resultsHide 45 results

DeepSeek-V4.1-Flash: common questions

Who makes DeepSeek-V4.1-Flash?

DeepSeek-V4.1-Flash is made by DeepSeek.

When was DeepSeek-V4.1-Flash released?

DeepSeek-V4.1-Flash was released on Sep 10, 2026, according to Artificial Analysis.

What is DeepSeek-V4.1-Flash good at?

DeepSeek-V4.1-Flash is strong in long context; capable in reasoning and coding; and behind the leaders in agentic tasks, multimodal tasks, math, factuality, and instruction following. Too few results yet to rate safety or multilingual tasks.

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, according to Artificial Analysis. We track its price at 5 providers. At a mix of three input tokens to one output token, it is cheaper than 56% of the 322 priced models we track.

How many benchmarks has DeepSeek-V4.1-Flash been tested on?

We track 110 results for DeepSeek-V4.1-Flash on 53 benchmarks from 18 sources, 49 of them independently verified. The latest was recorded on Oct 27, 2026.

Which API features does DeepSeek-V4.1-Flash support?

OpenRouter lists tool calling, structured outputs, and reasoning for DeepSeek-V4.1-Flash.

About this record

Where DeepSeek-V4.1-Flash's numbers come from, and every name it appears under.

Tracked since
Sep 9, 2026
Newest source mention
Oct 25, 2026

Where the results come from

Verification: 110 scores · 49 independently verified · 23 aggregator-attributed · 12 vendor cross-reference · 26 vendor-reported. How these tiers are assigned

From 18 sources on 10 sites. arcprize.org supplies 24 of them; the 49 independently verified results come from 8 sites. Bars are coloured by trust tier.

  • arcprize.org24
  • artificialanalysis.ai24
  • api.llm-stats.com22
  • huggingface.co16
  • livebench.ai7
  • epoch.ai6
  • datasets-server.huggingface.co5
  • raw.githubusercontent.com4
  • lmarena.ai1
  • simple-bench.com1

Also known as

How our sources name DeepSeek-V4.1-Flash at each reasoning setting.

SettingShort formAPI id
lowdeepseek v4.1 flash (low) deepseek v4.1 flash 1017 (low)—
highdeepseek v4.1 flash (high) deepseek v4.1 flash 1017 (high)—
maxdeepseek v4.1 flash (max) deepseek v4.1 flash 1017 (max)deepseek-v4.1-flash-20260910 (max) deepseek-v4.1-flash-20261017 (max) deepseek-v4.1-flash-max
Also listed asdeepseek 4.1 flashdeepseek flash v4.1deepseek v4-1 flashdeepseek v4.1 flash (non-reasoning)deepseek v4.1 flash (none)deepseek v4.1-flash-1017deepseek-ai/deepseek-v4.1-flashdeepseek-v4-1-flash-1017deepseek-v4-1-flash-non-reasoningdeepseek-v4.1-flash-20260910deepseek-v4.1-flash-20261017deepseek-v4.1-flash:batchv4.1-deepseek flashdeep seek v4.1 flash