llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeModelsDeepSeek-V2.5

DeepSeek model product

DeepSeek-V2.5

DeepSeek-V2.5 is an upgraded version that combines DeepSeek-V2-Chat and DeepSeek-Coder-V2-Instruct, integrating general and coding abilities. It better aligns with human preferences and has been optimized in various aspects, including writing and instruction following.

Updated Aug 10, 2026. Default version: DeepSeek-V2.5

Compare
LLMBoard score69.5DeepSeek-V2.5
Coverage60%6 benchmark families
Context window8.2KTokens
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-V2.5
Released
May 8, 2024
Knowledge cutoff
Unknown
Parameters
236B
Context window
8.2K
Max output
8.2K
Inputs
text
Outputs
text
Open weights
No
License
deepseek

Capability profile

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

DeepSeek-V2.5 category scores

Benchmark results

Published benchmark records for the scored version DeepSeek-V2.5.

15 rows
Columns

Show columns

Aider0.714100.0%CAug 7, 2026
DS-Arena-Code0.611100.0%CAug 7, 2026
DS-FIM-Eval0.811100.0%CAug 7, 2026
LiveCodeBench(01-09)0.411100.0%CAug 7, 2026
AlignBench0.82466.7%CAug 7, 2026
HumanEval-Mul0.7220.0%CAug 7, 2026
AlpacaEval 2.00.53433.3%CAug 7, 2026
MT-Bench0.931281.8%CAug 7, 2026
BBH0.851263.6%CAug 7, 2026
Arena Hard0.882672.0%CAug 7, 2026
GSM8k1.0114878.7%CAug 7, 2026
HumanEval0.9187677.3%CAug 7, 2026
MATH0.7297160.0%CAug 7, 2026
MMLU0.86010040.4%CAug 7, 2026
SWE-Bench Verified0.21031041.0%CAug 7, 2026

Arena results

Preference and agent-evaluation signals from published Arena datasets.

56 rows
Columns

Show columns

textjapanese1871190.5479N/AAug 6, 2026
text style controljapanese1881227.6479N/AAug 6, 2026
textfrench1901287.7222N/AAug 6, 2026
textkorean1901209.6344N/AAug 6, 2026
text style controlkorean1901254.8344N/AAug 6, 2026
text style controlfrench2021324.7222N/AAug 6, 2026
textspanish2131247.2293N/AAug 6, 2026
textchinese2171280.42,242N/AAug 6, 2026
text style controlchinese2191329.62,242N/AAug 6, 2026
text style controlcoding2191368.44,252N/AAug 6, 2026
text style controlspanish2201283.0293N/AAug 6, 2026
textgerman2211226.3563N/AAug 6, 2026
text style controlgerman2231261.1563N/AAug 6, 2026
textcoding2271301.84,252N/AAug 6, 2026
text style controlindustry legal and government2291330.21,518N/AAug 6, 2026
textindustry software and it services2321296.16,692N/AAug 6, 2026
text style controlindustry software and it services2321349.86,692N/AAug 6, 2026
textexpert2331239.01,523N/AAug 6, 2026
textindustry mathematical2331273.53,039N/AAug 6, 2026
textindustry legal and government2341282.21,518N/AAug 6, 2026
textindustry medicine and healthcare2351258.51,214N/AAug 6, 2026
textmath2351270.73,649N/AAug 6, 2026
texthard prompts2361270.76,869N/AAug 6, 2026
text style controlexpert2361294.81,523N/AAug 6, 2026
text style controlmath2371287.53,649N/AAug 6, 2026
text style controlhard prompts2381321.56,869N/AAug 6, 2026
text style controlindustry mathematical2381295.73,039N/AAug 6, 2026
texthard prompts english2401281.54,055N/AAug 6, 2026
textlonger query2411273.33,499N/AAug 6, 2026
text style controlhard prompts english2411330.54,055N/AAug 6, 2026
text style controlnon english2411290.011,727N/AAug 6, 2026
textindustry business and management and financial operations2421247.92,825N/AAug 6, 2026
textnon english2431253.011,727N/AAug 6, 2026
textrussian2431251.93,519N/AAug 6, 2026
text style controllonger query2431314.23,499N/AAug 6, 2026
text style controlindustry business and management and financial operations2441300.02,825N/AAug 6, 2026
text style controlrussian2441290.43,519N/AAug 6, 2026
text style controlindustry medicine and healthcare2451313.61,214N/AAug 6, 2026
textindustry life and physical and social science2461273.64,318N/AAug 6, 2026
textinstruction following2461249.510,175N/AAug 6, 2026
text style controlinstruction following2471291.510,175N/AAug 6, 2026
textexclude ties2481193.215,337N/AAug 6, 2026
text style controloverall2481307.224,572N/AAug 6, 2026
textoverall2491271.424,572N/AAug 6, 2026
text style controlindustry life and physical and social science2491320.84,318N/AAug 6, 2026
text style controlmulti turn2491292.64,473N/AAug 6, 2026
textenglish2501284.412,845N/AAug 6, 2026
textmulti turn2501260.74,473N/AAug 6, 2026
text style controlindustry entertainment and sports and media2501264.03,664N/AAug 6, 2026
text style controlindustry writing and literature and language2511283.56,613N/AAug 6, 2026
textindustry entertainment and sports and media2521230.53,664N/AAug 6, 2026
textcreative writing2531237.13,450N/AAug 6, 2026
text style controlcreative writing2531264.93,450N/AAug 6, 2026
text style controlexclude ties2531243.615,337N/AAug 6, 2026
text style controlenglish2581318.512,845N/AAug 6, 2026
textindustry writing and literature and language2591246.86,613N/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
0
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-V2.5 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-V2.5May 8, 202469.5236B8.2K8.2KNodeepseek

What is DeepSeek-V2.5?

A concise description based on the published model registry.

DeepSeek-V2.5 is an upgraded version that combines DeepSeek-V2-Chat and DeepSeek-Coder-V2-Instruct, integrating general and coding abilities. It better aligns with human preferences and has been optimized in various aspects, including writing and instruction following.

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-V2.5 vs nearby models

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

DeepSeek-V2.5vsKimi K2.5DeepSeek-V2.5vsGPT-5DeepSeek-V2.5vsLlama 3.1 405BDeepSeek-V2.5vsClaude Opus 4.6DeepSeek-V2.5vsGPT-5.2DeepSeek-V2.5vsMiMo V2.5 Pro

Models similar to DeepSeek-V2.5

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

#28+2.2
DE

DeepSeek-V4-Flash

DeepSeek

71.7 LLMBoard

DetailsCompare
#18+6.0
DE

DeepSeek-V3.2

DeepSeek

75.5 LLMBoard

DetailsCompare
#12+10.6
DE

DeepSeek-V4-Pro

DeepSeek

80.1 LLMBoard

DetailsCompare
#101-25.4
DE

DeepSeek-R1

DeepSeek

44.1 LLMBoard

DetailsCompare
#119-31.4
DE

DeepSeek-V3

DeepSeek

38.1 LLMBoard

DetailsCompare
#126-33.5
DE

DeepSeek-V3.1

DeepSeek

36.0 LLMBoard

DetailsCompare

FAQ

Common questions about DeepSeek-V2.5.

When was DeepSeek-V2.5 released?

DeepSeek-V2.5's default version was released on May 8, 2024.

How much does DeepSeek-V2.5 cost?

No official standard PAYG price is currently available for DeepSeek-V2.5.

Who created DeepSeek-V2.5?

DeepSeek-V2.5 is published under DeepSeek in the model registry.

What is the context window for DeepSeek-V2.5?

The default version has a 8.2K token context window.

Is DeepSeek-V2.5 open weight?

No. The default version is not marked as having publicly available weights.

How many API providers offer DeepSeek-V2.5?

No published provider offering is currently linked to the default version.

What models should I compare DeepSeek-V2.5 with?

Nearby ranked alternatives include Kimi K2.5, GPT-5, Llama 3.1 405B.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai