Enter opens the highlighted result. Esc closes.

DeepSeek V4 Pro 0423 vs Kimi K2.5

DeepSeek V4 Pro 0423 leads Kimi K2.5 on coding, 38.7 to 38.6 on the Vibe Coding Index, at about the same output price ($2.976 vs $2.85 per million). Updated October 22, 2026; 66 benchmarks have a result for both.

Replace DeepSeek V4 Pro 0423 Replace Kimi K2.5
DeepSeek V4 Pro 0423DeepSeek
Coding
38.7
Overall
50.0
Output $/M
$2.976
47
Kimi K2.5Moonshot AI
Coding
38.6
Overall
44.8
Output $/M
$2.85
19
vs boards won of 66 shared

DeepSeek V4 Pro 0423 or Kimi K2.5: which to pick

DeepSeek V4 Pro 0423 and Kimi K2.5 are level for coding, 38.7 against 38.6 on the Vibe Coding Index, so price and context decide.

Pick Kimi K2.5 for output tokens 1.0x cheaper ($2.85 vs $2.976 per million) and a lower cost per solved task ($0.24 against $0.57). Pick DeepSeek V4 Pro 0423 for a larger context window (1.05M against 262K tokens) and work measured by LMArena WebDev or AA Coding Index, where it leads.

Model ADeepSeek V4 Pro 0423DeepSeek38.7/1005 independent benchmarks$0.57 per solved taskModel BKimi K2.5Moonshot AI38.6/1003 independent benchmarks$0.24 per solved task
Shared results
66
DeepSeek V4 Pro 0423 only
18
Kimi K2.5 only
33
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
DeepSeek V4 Pro 0423 38.7 from 5 benchmarks, Kimi K2.5 38.6 from 3 benchmarks.
Widest category
Agentic work, 15.6 points apart on average over 7 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): DeepSeek V4 Pro 0423 49.9% against Kimi K2.5 17.5%, 32.4 points apart on a 0 to 100 scale.
Output price
DeepSeek V4 Pro 0423 $2.976 against Kimi K2.5 $2.85 per million output tokens, 1.0x apart.
Context window
DeepSeek V4 Pro 0423 takes 1.05M tokens, Kimi K2.5 262K.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryDeepSeek V4 Pro 0423BarsKimi K2.5Ties
Coding112
Intelligence20
Agentic work43
Reasoning32
Math02
Knowledge82
Web design35
General163
All shared boards4719
CodingLike-for-likeSix of these boards feed the Vibe Coding Index112
Coding benchmarks for DeepSeek V4 Pro 0423 and Kimi K2.5
BenchmarkDeepSeek V4 Pro 0423Kimi K2.5Leader
SWE-bench VerifiedNot measured70.8%Only Kimi K2.5 measured
SWE-bench MultilingualNot measured67.3%Only Kimi K2.5 measured
LMArena WebDev14471437DeepSeek V4 Pro 0423
AA Coding Index59.446.8DeepSeek V4 Pro 0423
LiveCodeBench (Vals AI)87.5%83.9%DeepSeek V4 Pro 0423
SWE-bench Verified (Vals AI)77.4%70.0%DeepSeek V4 Pro 0423
Terminal-Bench 4 (Vals AI)11.1%Not measuredOnly DeepSeek V4 Pro 0423 measured
Vibe Code Bench (Vals AI)49.9%17.5%DeepSeek V4 Pro 0423
Code migration (Vals AI)26.2%7.0%DeepSeek V4 Pro 0423
ProgramBench (Vals AI)0.0%Not measuredOnly DeepSeek V4 Pro 0423 measured
Terminal-Bench 2.1 (Vals AI)50.2%42.0%DeepSeek V4 Pro 0423
SkillsBench (Vals AI)51.3%Not measuredOnly DeepSeek V4 Pro 0423 measured
Terminal-Bench Hard (Artificial Analysis run)Not measured34.8%Only Kimi K2.5 measured
Terminal-BenchNot measured43.2%Only Kimi K2.5 measured
FrontierCode17.6%Not measuredOnly DeepSeek V4 Pro 0423 measured
WeirdML48.9%45.6%DeepSeek V4 Pro 0423
ALE-Bench1006.1821.6DeepSeek V4 Pro 0423
LMArena Text: coding14701483Kimi K2.5
LMArena Text: software and IT14691471Kimi K2.5
LMArena WebDev: React14411429DeepSeek V4 Pro 0423
LMArena WebDev: HTML14351426DeepSeek V4 Pro 0423
LMArena WebDev: from an imageNot measured1439Only Kimi K2.5 measured
Neither model has a result on the other 17 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for DeepSeek V4 Pro 0423 and Kimi K2.5
BenchmarkDeepSeek V4 Pro 0423Kimi K2.5Leader
AA Intelligence Index30.423.5DeepSeek V4 Pro 0423
Epoch Capabilities Index149.1148.1DeepSeek V4 Pro 0423
Agentic workLike-for-likePublic boards, shown as published, kept out of the index43
Agentic work benchmarks for DeepSeek V4 Pro 0423 and Kimi K2.5
BenchmarkDeepSeek V4 Pro 0423Kimi K2.5Leader
AA Agentic Index26.3Not measuredOnly DeepSeek V4 Pro 0423 measured
Design Arena: agents building web apps10001136Kimi K2.5
Design Arena: agents building full-stack apps9481114Kimi K2.5
Design Arena: agents building mobile appsNot measured1147Only Kimi K2.5 measured
Design Arena: agents building Android appsNot measured1118Only Kimi K2.5 measured
Design Arena: agents building Godot games10591219Kimi K2.5
Finance Agent (Vals AI)44.1%35.8%DeepSeek V4 Pro 0423
Tax Agent Bench (Vals AI)25.2%Not measuredOnly DeepSeek V4 Pro 0423 measured
Legal Research Bench (Vals AI)23.1%15.9%DeepSeek V4 Pro 0423
Excel Modeling Benchmark (Vals AI)51.6%28.5%DeepSeek V4 Pro 0423
τ²-Bench TelecomNot measured95.9%Only Kimi K2.5 measured
OSWorldNot measured63.3%Only Kimi K2.5 measured
Vending-Bench 2$3,285$1,198DeepSeek V4 Pro 0423
Neither model has a result on the other 28 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index32
Reasoning benchmarks for DeepSeek V4 Pro 0423 and Kimi K2.5
BenchmarkDeepSeek V4 Pro 0423Kimi K2.5Leader
GPQA Diamond89.4%84.1%DeepSeek V4 Pro 0423
MMMU Pro (Vals AI)Not measured84.3%Only Kimi K2.5 measured
GPQA Diamond (Artificial Analysis run)Not measured87.9%Only Kimi K2.5 measured
Humanity's Last Exam (Artificial Analysis run)Not measured30.7%Only Kimi K2.5 measured
GPQA Diamond (Epoch AI run)90.9%Not measuredOnly DeepSeek V4 Pro 0423 measured
ARC-AGINot measured65.3%Only Kimi K2.5 measured
ARC-AGI-2Not measured11.8%Only Kimi K2.5 measured
SimpleBenchNot measured46.8%Only Kimi K2.5 measured
Chess puzzles19.0%Not measuredOnly DeepSeek V4 Pro 0423 measured
Mystery game puzzles17.0%Not measuredOnly DeepSeek V4 Pro 0423 measured
LMArena Text: hard prompts14611453DeepSeek V4 Pro 0423
LMArena Text: expert14631466Kimi K2.5
LMArena Text: hard prompts in English14661460DeepSeek V4 Pro 0423
LMArena Vision: diagramsNot measured1282Only Kimi K2.5 measured
LMArena Vision: homeworkNot measured1297Only Kimi K2.5 measured
Kagi LLM Benchmark53.5%78.5%Kimi K2.5
Neither model has a result on the other 6 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for DeepSeek V4 Pro 0423 and Kimi K2.5
BenchmarkDeepSeek V4 Pro 0423Kimi K2.5Leader
AIMENot measured95.6%Only Kimi K2.5 measured
OTIS Mock AIME96.7%Not measuredOnly DeepSeek V4 Pro 0423 measured
FrontierMath, tiers 1 to 345.3%Not measuredOnly DeepSeek V4 Pro 0423 measured
FrontierMath, tier 42.4%Not measuredOnly DeepSeek V4 Pro 0423 measured
ProofBench16.0%Not measuredOnly DeepSeek V4 Pro 0423 measured
LMArena Text: math14431470Kimi K2.5
LMArena Text: mathematical work14551473Kimi K2.5
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index82
Knowledge benchmarks for DeepSeek V4 Pro 0423 and Kimi K2.5
BenchmarkDeepSeek V4 Pro 0423Kimi K2.5Leader
MMLU Pro87.3%85.9%DeepSeek V4 Pro 0423
LegalBench (Vals AI)80.3%Not measuredOnly DeepSeek V4 Pro 0423 measured
CaseLaw (Vals AI)59.4%58.7%DeepSeek V4 Pro 0423
CorpFin (Vals AI)61.4%68.3%Kimi K2.5
TaxEval (Vals AI)72.1%74.2%Kimi K2.5
MortgageTax (Vals AI)Not measured66.5%Only Kimi K2.5 measured
MedQA (Vals AI)Not measured94.4%Only Kimi K2.5 measured
MedCode (Vals AI)40.5%39.3%DeepSeek V4 Pro 0423
SimpleQA Verified47.0%34.3%DeepSeek V4 Pro 0423
LMArena Text: science14701459DeepSeek V4 Pro 0423
LMArena Text: medicine and healthcare14661448DeepSeek V4 Pro 0423
LMArena Text: legal and government14641441DeepSeek V4 Pro 0423
LMArena Text: business and finance14431427DeepSeek V4 Pro 0423
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index35
Web design benchmarks for DeepSeek V4 Pro 0423 and Kimi K2.5
BenchmarkDeepSeek V4 Pro 0423Kimi K2.5Leader
Design Arena: websites12341248Kimi K2.5
Design Arena: UI components12191234Kimi K2.5
Design Arena: game dev12281206DeepSeek V4 Pro 0423
Design Arena: data viz12031219Kimi K2.5
Design Arena: 3D12531204DeepSeek V4 Pro 0423
Design Arena: all code categories12401236DeepSeek V4 Pro 0423
Design Arena: SVG11391149Kimi K2.5
Design Arena: ASCII art11491166Kimi K2.5
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index163
General benchmarks for DeepSeek V4 Pro 0423 and Kimi K2.5
BenchmarkDeepSeek V4 Pro 0423Kimi K2.5Leader
Public Benefits Bench (Vals AI)62.9%Not measuredOnly DeepSeek V4 Pro 0423 measured
SAGE (Vals AI)Not measured49.9%Only Kimi K2.5 measured
MedScribe (Vals AI)75.1%76.4%Kimi K2.5
IFBenchNot measured70.2%Only Kimi K2.5 measured
Long context reasoning (AA-LCR)Not measured78.0%Only Kimi K2.5 measured
DTBench90.7%Not measuredOnly DeepSeek V4 Pro 0423 measured
LMCA41.2%Not measuredOnly DeepSeek V4 Pro 0423 measured
Fiction.LiveBenchNot measured86.1%Only Kimi K2.5 measured
CL-benchNot measured19.3%Only Kimi K2.5 measured
Surface Evolver Bench40.0%Not measuredOnly DeepSeek V4 Pro 0423 measured
CL-bench Life13.5%13.2%DeepSeek V4 Pro 0423
LMArena Text14511445DeepSeek V4 Pro 0423
LMArena Text: instruction following14451431DeepSeek V4 Pro 0423
LMArena Text: creative writing14461422DeepSeek V4 Pro 0423
LMArena VisionNot measured1269Only Kimi K2.5 measured
LMArena Text: English14591453DeepSeek V4 Pro 0423
LMArena Text: non-English14391433DeepSeek V4 Pro 0423
LMArena Text: Chinese14811494Kimi K2.5
LMArena Text: German14581441DeepSeek V4 Pro 0423
LMArena Text: Japanese14461423DeepSeek V4 Pro 0423
LMArena Text: Korean14091412Kimi K2.5
LMArena Text: French14731454DeepSeek V4 Pro 0423
LMArena Text: Spanish14571450DeepSeek V4 Pro 0423
LMArena Text: Russian14531435DeepSeek V4 Pro 0423
LMArena Text: Polish14611445DeepSeek V4 Pro 0423
LMArena Text: multi-turn14681444DeepSeek V4 Pro 0423
LMArena Text: longer queries14581445DeepSeek V4 Pro 0423
LMArena Text: writing and language14471425DeepSeek V4 Pro 0423
LMArena Text: entertainment and media14311417DeepSeek V4 Pro 0423
LMArena Vision: EnglishNot measured1274Only Kimi K2.5 measured
LMArena Vision: ChineseNot measured1328Only Kimi K2.5 measured
LMArena Vision: reading textNot measured1276Only Kimi K2.5 measured
LMArena Vision: humorNot measured1284Only Kimi K2.5 measured
LMArena Vision: captioningNot measured1257Only Kimi K2.5 measured
LMArena Vision: recognizing thingsNot measured1285Only Kimi K2.5 measured
LMArena Vision: creative writingNot measured1269Only Kimi K2.5 measured
Neither model has a result on the other 10 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for DeepSeek V4 Pro 0423 and Kimi K2.5
BenchmarkDeepSeek V4 Pro 0423Kimi K2.5Leader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

DeepSeek V4 Pro 0423$0.0028Fits in one request
Kimi K2.5$0.0020Fits in one request

Kimi K2.5 costs less

Repository review

50K input + 3K output tokens

DeepSeek V4 Pro 0423$0.077Fits in one request
Kimi K2.5$0.035Fits in one request

Kimi K2.5 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4 Pro 0423$0.301Fits in one request
Kimi K2.5$0.135Fits in one request

Kimi K2.5 costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of DeepSeek V4 Pro 0423 and Kimi K2.5
SpecificationDeepSeek V4 Pro 0423MoonshotAI Kimi K2.5
MakerDeepSeekMoonshot AI
WeightsOpen weightsOpen weights
ReleasedApril 24, 2026January 27, 2026
Context window1.05M262K
Input $/Mtok$1.358$0.532
Output $/Mtok$2.976$2.85
Coding index38.738.6
Independent measurements behind it53
Cost per solved task$0.57$0.24
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V4 Pro 0423 better than Kimi K2.5 for coding?

DeepSeek V4 Pro 0423 and Kimi K2.5 are level for coding, 38.7 against 38.6 on the Vibe Coding Index, so price and context decide. Pick Kimi K2.5 for output tokens 1.0x cheaper ($2.85 vs $2.976 per million) and a lower cost per solved task ($0.24 against $0.57). Pick DeepSeek V4 Pro 0423 for a larger context window (1.05M against 262K tokens) and work measured by LMArena WebDev or AA Coding Index, where it leads.

Which is cheaper, DeepSeek V4 Pro 0423 or Kimi K2.5?

DeepSeek V4 Pro 0423 costs $1.358 per million input tokens and $2.976 per million output tokens; Kimi K2.5 costs $0.532 and $2.85. Per solved SWE-bench Verified task, DeepSeek V4 Pro 0423 costs $0.57 and Kimi K2.5 $0.24.

Which has the bigger context window, DeepSeek V4 Pro 0423 or Kimi K2.5?

DeepSeek V4 Pro 0423, with 1.05M tokens against 262K.

How many benchmarks is the DeepSeek V4 Pro 0423 vs Kimi K2.5 comparison based on?

66 benchmarks have a result for both models as of October 22, 2026. DeepSeek V4 Pro 0423's coding index rests on 5 independent benchmarks and Kimi K2.5's on 3. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/deepseek-v4-pro-vs-kimi-k2-5" title="DeepSeek V4 Pro 0423 vs Kimi K2.5, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/deepseek-v4-pro-vs-kimi-k2-5">LLMBenchmark.ai, DeepSeek V4 Pro 0423 vs Kimi K2.5</a></p>