llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeModelsDeepSeek-V3.1

DeepSeek model product

DeepSeek-V3.1

DeepSeek-V3.1 is a hybrid model supporting both thinking and non-thinking modes through different chat templates. Built on DeepSeek-V3.1-Base with a two-phase long context extension (32K phase: 630B tokens, 128K phase: 209B tokens), it features 671B total parameters with 37B activated. Key improvements include smarter tool calling through post-training optimization, higher thinking efficiency achieving comparable quality to DeepSeek-R1-0528 while responding more quickly, and UE8M0 FP8 scale data format for model weights and activations. The model excels in both reasoning tasks (thinking mode) and practical applications (non-thinking mode), with particularly strong performance in code agent tasks, math competitions, and search-based problem solving.

Updated Aug 10, 2026. Default version: DeepSeek-V3.1

Compare
LLMBoard score36.0DeepSeek-V3.1
Coverage100%10 benchmark families
Context window131.1KTokens
Official input priceN/AOfficial price unavailable

On this page

  • Specification
  • Capability
  • Benchmarks
  • Arena
  • Pricing
  • Versions
  • About
  • Compare
  • Similar models
  • FAQ

Model specification

Structured fields from the published default version.

Version
DeepSeek-V3.1
Released
Jan 10, 2025
Knowledge cutoff
Unknown
Parameters
671B
Context window
131.1K
Max output
8.2K
Inputs
text
Outputs
text
Open weights
Yes
License
MIT

Capability profile

This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.

DeepSeek-V3.1 category scores

Benchmark results

Published benchmark records for the scored version DeepSeek-V3.1.

16 rows
Columns

Show columns

SimpleQA0.934695.6%CAug 7, 2026
Aider-Polyglot0.782266.7%CAug 7, 2026
BrowseComp-zh0.5101325.0%CAug 7, 2026
CodeForces0.7121626.7%CAug 7, 2026
Terminal-Bench0.3182529.2%CAug 7, 2026
MMLU-Redux0.9214857.5%CAug 7, 2026
SWE-bench Multilingual0.5293415.2%CAug 7, 2026
MMLU-Pro0.83012977.3%CAug 7, 2026
HMMT 20250.332333.1%CAug 7, 2026
LiveCodeBench0.6367351.4%CAug 7, 2026
AIME 20240.7435319.2%CAug 7, 2026
BrowseComp0.355585.3%CAug 7, 2026
Humanity's Last Exam0.2689226.4%CAug 7, 2026
SWE-Bench Verified0.77110432.0%CAug 7, 2026
AIME 20250.510011412.4%CAug 7, 2026
GPQA0.710823353.9%CAug 7, 2026

Arena results

Preference and agent-evaluation signals from published Arena datasets.

58 rows
Columns

Show columns

textfrench541457.3246N/AAug 6, 2026
text style controlfrench841452.8246N/AAug 6, 2026
text style controljapanese871376.0308N/AAug 6, 2026
textjapanese891367.8308N/AAug 6, 2026
textgerman931409.4374N/AAug 6, 2026
textindustry medicine and healthcare931432.4935N/AAug 6, 2026
textpolish931409.5992N/AAug 6, 2026
textindustry writing and literature and language961397.83,459N/AAug 6, 2026
textchinese971458.01,110N/AAug 6, 2026
textenglish971430.76,773N/AAug 6, 2026
textindustry life and physical and social science981433.42,313N/AAug 6, 2026
textoverall991419.614,942N/AAug 6, 2026
text style controlchinese1001463.01,110N/AAug 6, 2026
textmath1011420.7992N/AAug 6, 2026
textspanish1011412.0362N/AAug 6, 2026
textexclude ties1021405.810,654N/AAug 6, 2026
textcreative writing1041388.22,000N/AAug 6, 2026
text style controlgerman1061405.5374N/AAug 6, 2026
textindustry business and management and financial operations1071405.92,518N/AAug 6, 2026
textindustry entertainment and sports and media1071379.82,747N/AAug 6, 2026
text style controlindustry writing and literature and language1071399.03,459N/AAug 6, 2026
text style controlspanish1071405.2362N/AAug 6, 2026
text style controlindustry business and management and financial operations1101414.62,518N/AAug 6, 2026
textindustry legal and government1111418.31,011N/AAug 6, 2026
text style controloverall1111417.614,942N/AAug 6, 2026
text style controlcreative writing1111389.02,000N/AAug 6, 2026
text style controlexclude ties1121404.710,654N/AAug 6, 2026
text style controlmath1121414.5992N/AAug 6, 2026
text style controlpolish1121400.1992N/AAug 6, 2026
textindustry mathematical1131415.3839N/AAug 6, 2026
text style controlindustry life and physical and social science1131434.62,313N/AAug 6, 2026
textnon english1141398.88,169N/AAug 6, 2026
textrussian1141400.7790N/AAug 6, 2026
text style controlenglish1141428.16,773N/AAug 6, 2026
text style controlindustry medicine and healthcare1151434.9935N/AAug 6, 2026
text style controlinstruction following1161403.53,689N/AAug 6, 2026
textindustry software and it services1171428.34,783N/AAug 6, 2026
text style controlhard prompts1171433.06,801N/AAug 6, 2026
texthard prompts1191415.36,801N/AAug 6, 2026
text style controllonger query1191421.33,210N/AAug 6, 2026
textinstruction following1201389.43,689N/AAug 6, 2026
text style controlindustry legal and government1211421.01,011N/AAug 6, 2026
text style controlindustry software and it services1211445.44,783N/AAug 6, 2026
text style controlnon english1211397.68,169N/AAug 6, 2026
texthard prompts english1221418.63,357N/AAug 6, 2026
text style controlindustry entertainment and sports and media1221378.72,747N/AAug 6, 2026
text style controlexpert1231426.5726N/AAug 6, 2026
textlonger query1241400.53,210N/AAug 6, 2026
text style controlrussian1241401.2790N/AAug 6, 2026
textexpert1251402.8726N/AAug 6, 2026
text style controlhard prompts english1251435.53,357N/AAug 6, 2026
text style controlindustry mathematical1251414.0839N/AAug 6, 2026
textcoding1281415.22,624N/AAug 6, 2026
textkorean1281331.9472N/AAug 6, 2026
textmulti turn1281401.72,482N/AAug 6, 2026
text style controlmulti turn1301406.32,482N/AAug 6, 2026
text style controlcoding1321448.32,624N/AAug 6, 2026
text style controlkorean1321338.2472N/AAug 6, 2026

Pricing

Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.

Official API
N/A
Official provider
N/A
Lowest third-party
N/A
Tracked offerings
1
No published price snapshot

The default version has no provider offering with current input or output token prices.

Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.

DeepSeek-V3.1 versions

All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.

1 rows
Columns

Show columns

DeepSeek-V3.1Jan 10, 202536.0671B131.1K8.2KYesMIT

What is DeepSeek-V3.1?

A concise description based on the published model registry.

DeepSeek-V3.1 is a hybrid model supporting both thinking and non-thinking modes through different chat templates. Built on DeepSeek-V3.1-Base with a two-phase long context extension (32K phase: 630B tokens, 128K phase: 209B tokens), it features 671B total parameters with 37B activated. Key improvements include smarter tool calling through post-training optimization, higher thinking efficiency achieving comparable quality to DeepSeek-R1-0528 while responding more quickly, and UE8M0 FP8 scale data format for model weights and activations. The model excels in both reasoning tasks (thinking mode) and practical applications (non-thinking mode), with particularly strong performance in code agent tasks, math competitions, and search-based problem solving.

Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.

Data snapshot: 2026-08-07. Editorial model content is not available in the backend.

DeepSeek-V3.1 vs nearby models

Open a comparison with the three ranked models immediately above and below this model.

DeepSeek-V3.1vsNova MicroDeepSeek-V3.1vsQwen3.5 9BDeepSeek-V3.1vsQwen3 VL 32BDeepSeek-V3.1vsCommand R+DeepSeek-V3.1vsMistral Small 3 24BDeepSeek-V3.1vsClaude Haiku 3

Models similar to DeepSeek-V3.1

Recommendations prioritize the same model type and family, then the closest published LLMBoard score.

#119+2.1
DE

DeepSeek-V3

DeepSeek

38.1 LLMBoard

DetailsCompare
#101+8.1
DE

DeepSeek-R1

DeepSeek

44.1 LLMBoard

DetailsCompare
#32+33.5
DE

DeepSeek-V2.5

DeepSeek

69.5 LLMBoard

DetailsCompare
#28+35.7
DE

DeepSeek-V4-Flash

DeepSeek

71.7 LLMBoard

DetailsCompare
#18+39.5
DE

DeepSeek-V3.2

DeepSeek

75.5 LLMBoard

DetailsCompare
#12+44.1
DE

DeepSeek-V4-Pro

DeepSeek

80.1 LLMBoard

DetailsCompare

FAQ

Common questions about DeepSeek-V3.1.

When was DeepSeek-V3.1 released?

DeepSeek-V3.1's default version was released on Jan 10, 2025.

How much does DeepSeek-V3.1 cost?

No official standard PAYG price is currently available for DeepSeek-V3.1.

Who created DeepSeek-V3.1?

DeepSeek-V3.1 is published under DeepSeek in the model registry.

What is the context window for DeepSeek-V3.1?

The default version has a 131.1K token context window.

Is DeepSeek-V3.1 open weight?

Yes. The default version is marked as open weight under MIT.

How many API providers offer DeepSeek-V3.1?

1 published provider offerings are linked to the default version.

What models should I compare DeepSeek-V3.1 with?

Nearby ranked alternatives include Nova Micro, Qwen3.5 9B, Qwen3 VL 32B.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai