Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d269Policy30d65, 14 pendingSources1,014

DeepSeek-V3.1

DeepSeekReleased Aug 21, 2025Current

Score basis

DeepSeek-V3.1 is behind the leaders in factuality, long context, reasoning, coding, and agentic tasks. Too few results yet to rate safety, math, multimodal tasks, multilingual tasks, or instruction following.

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.

Limited5 capabilities
  1. Factuality−26.2%3 of 4
  2. Long Context−33.1%1 of 3
  3. Reasoning−37.3%4 of 6
  4. Coding−38.0%4 of 10
  5. Agentic−45.8%2 of 7
Not rated5 capabilities

Too few results yet to rate Safety, Math, Multimodal, Multilingual or Instruction Following.

Price

$0.56input$1.68outputper million tokens

From Artificial Analysis · 2 providers tracked · All prices

Costs more than 54% of 331 priced models · 3:1 input-to-output blend, log scale

Evidence

67results on39benchmarks

  • 8 independently verified
  • 30 aggregator
  • 24 vendor-reported
  • 5 cross-referenced

From 9 sources · latest Oct 8, 2026 · How verification works

Research

10 papers reference DeepSeek-V3.1

DeepSeek-V3.1 benchmark results

67 results on 39 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Factuality

LimitedFull factuality ranking

26.2% behind the leader3 of 4 ranked benchmarks measured

Show 2 more factuality resultsHide 2 factuality results

Long Context

LimitedFull long context ranking

33.1% behind the leader1 of 3 ranked benchmarks measured

Show 1 more long context resultHide 1 long context result

Reasoning

LimitedFull reasoning ranking

37.3% behind the leader4 of 6 ranked benchmarks measured

Show 6 more reasoning resultsHide 6 reasoning results

Coding

LimitedFull coding ranking

38.0% behind the leader4 of 10 ranked benchmarks measured

Show 4 more coding resultsHide 4 coding results

Agentic

LimitedFull agentic ranking

45.8% behind the leader2 of 7 ranked benchmarks measured

Show 1 more agentic resultHide 1 agentic result

Instruction Following

Not enough dataFull instruction following ranking

0 of 3 ranked benchmarks measured

More results

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

Show 30 more resultsHide 30 results

DeepSeek-V3.1: common questions

Who makes DeepSeek-V3.1?

DeepSeek-V3.1 is made by DeepSeek.

When was DeepSeek-V3.1 released?

DeepSeek-V3.1 was released on Aug 21, 2025, according to Artificial Analysis.

What is DeepSeek-V3.1 good at?

DeepSeek-V3.1 is behind the leaders in factuality, long context, reasoning, coding, and agentic tasks. Too few results yet to rate safety, math, multimodal tasks, multilingual tasks, or instruction following.

How much does DeepSeek-V3.1 cost?

DeepSeek-V3.1 costs $0.56 per million input tokens and $1.68 per million output tokens, according to Artificial Analysis. We track its price at 2 providers. At a mix of three input tokens to one output token, it costs more than 54% of the 331 priced models we track.

How many benchmarks has DeepSeek-V3.1 been tested on?

We track 67 results for DeepSeek-V3.1 on 39 benchmarks from 9 sources, 8 of them independently verified. The latest was recorded on Oct 8, 2026.

About this record

Where DeepSeek-V3.1's numbers come from, and every name it appears under.

Tracked since
Apr 27, 2026
Newest source mention
Aug 23, 2026

Where the results come from

Verification: 67 scores · 8 independently verified · 30 aggregator-attributed · 5 vendor cross-reference · 24 vendor-reported. How these tiers are assigned

From 9 sources on 7 sites. Artificial Analysis supplies 30 of them; the 8 independently verified results come from 4 sites. Bars are coloured by trust tier.

  • artificialanalysis.ai30
  • huggingface.co15
  • api.llm-stats.com14
  • raw.githubusercontent.com4
  • matharena.ai2
  • labs.scale.com1
  • simple-bench.com1

Also known as

DeepSeek V3.1 (Non-reasoning)deepseek-v3p1deepseek-v3-1deepseek-v3-1-reasoningdeepseek-v3.1 (think)deepseek v3.1 (reasoning)deepseek-v3-1 (Non-Reasoning)DeepSeek V3.1-NonThinking (Non-Reasoning)deepseek-v3-1 (reasoning)deepseek v3.1-thinking