Enter opens the highlighted result. Esc closes.

Kimi K2.5 vs o4 Mini

Kimi K2.5 leads o4 Mini on the Artificial Analysis Intelligence Index, 23.5 to 16.7, and Kimi K2.5 costs 1.5x less per output token ($2.85 vs $4.40 per million). Updated October 22, 2026; 72 benchmarks have a result for both.

Replace Kimi K2.5 Replace o4 Mini
Kimi K2.5Moonshot AI
Coding
38.6
Overall
44.8
Output $/M
$2.85
69
o4 MiniOpenAI
Coding
No data
Overall
43.0
Output $/M
$4.40
3
vs boards won of 72 shared

Kimi K2.5 or o4 Mini: which to pick

Only Kimi K2.5 has a coding score, 38.6 on the Vibe Coding Index; o4 Mini has none, so the two cannot be ranked on coding.

Pick o4 Mini for work measured by WeirdML or TaxEval (Vals AI), where it leads.

Model AKimi K2.5Moonshot AI38.6/1003 independent benchmarks$0.24 per solved taskModel Bo4 MiniOpenAINot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
72
Kimi K2.5 only
27
o4 Mini only
26
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
Kimi K2.5 38.6 from 3 benchmarks, o4 Mini not benchmarked.
Widest category
Agentic work, 40.4 points apart on average over 1 shared benchmark.
Widest single gap
Design Arena: 3D: Kimi K2.5 1204 against o4 Mini 854, 43.8 points apart on a 0 to 100 scale.
Output price
Kimi K2.5 $2.85 against o4 Mini $4.40 per million output tokens, 1.5x apart.
Context window
Kimi K2.5 takes 262K tokens, o4 Mini 200K.

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
CategoryKimi K2.5Barso4 MiniTies
Coding51
Intelligence20
Agentic work10
Reasoning130
Math30
Knowledge92
Web design60
General300
All shared boards693
CodingLike-for-likeSix of these boards feed the Vibe Coding Index51
Coding benchmarks for Kimi K2.5 and o4 Mini
BenchmarkKimi K2.5o4 MiniLeader
Aider polyglotNot measured72.0%Only o4 Mini measured
SWE-bench Verified70.8%64.6%Kimi K2.5
SWE-bench MultimodalNot measured33.9%Only o4 Mini measured
SWE-bench Multilingual67.3%Not measuredOnly Kimi K2.5 measured
LMArena WebDev1437Not measuredOnly Kimi K2.5 measured
AA Coding Index46.8Not measuredOnly Kimi K2.5 measured
LiveCodeBench (Vals AI)83.9%82.2%Kimi K2.5
SWE-bench Verified (Vals AI)70.0%Not measuredOnly Kimi K2.5 measured
Vibe Code Bench (Vals AI)17.5%Not measuredOnly Kimi K2.5 measured
Code migration (Vals AI)7.0%Not measuredOnly Kimi K2.5 measured
Terminal-Bench 2.1 (Vals AI)42.0%Not measuredOnly Kimi K2.5 measured
LiveCodeBench (Artificial Analysis run)Not measured85.9%Only o4 Mini measured
Terminal-Bench Hard (Artificial Analysis run)34.8%15.2%Kimi K2.5
Terminal-Bench43.2%Not measuredOnly Kimi K2.5 measured
GSO-BenchNot measured3.6%Only o4 Mini measured
WeirdML45.6%52.6%o4 Mini
ALE-Bench821.6Not measuredOnly Kimi K2.5 measured
LMArena Text: coding14831368Kimi K2.5
LMArena Text: software and IT14711367Kimi K2.5
LMArena WebDev: React1429Not measuredOnly Kimi K2.5 measured
LMArena WebDev: HTML1426Not measuredOnly Kimi K2.5 measured
LMArena WebDev: from an image1439Not measuredOnly Kimi K2.5 measured
SciCodeNot measured10.8%Only o4 Mini measured
Neither model has a result on the other 16 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Kimi K2.5 and o4 Mini
BenchmarkKimi K2.5o4 MiniLeader
AA Intelligence Index23.516.7Kimi K2.5
Epoch Capabilities Index148.1145.5Kimi K2.5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for Kimi K2.5 and o4 Mini
BenchmarkKimi K2.5o4 MiniLeader
Design Arena: agents building web apps1136Not measuredOnly Kimi K2.5 measured
Design Arena: agents building full-stack apps1114Not measuredOnly Kimi K2.5 measured
Design Arena: agents building mobile apps1147Not measuredOnly Kimi K2.5 measured
Design Arena: agents building Android apps1118Not measuredOnly Kimi K2.5 measured
Design Arena: agents building Godot games1219Not measuredOnly Kimi K2.5 measured
Finance Agent (Vals AI)35.8%Not measuredOnly Kimi K2.5 measured
Legal Research Bench (Vals AI)15.9%Not measuredOnly Kimi K2.5 measured
Excel Modeling Benchmark (Vals AI)28.5%Not measuredOnly Kimi K2.5 measured
τ²-Bench Telecom95.9%55.6%Kimi K2.5
METR time horizonsNot measured63.9%Only o4 Mini measured
OSWorld63.3%Not measuredOnly Kimi K2.5 measured
Vending-Bench 2$1,198Not measuredOnly Kimi K2.5 measured
Berkeley Function Calling Leaderboard (BFCL)Not measured53.2%Only o4 Mini measured
Neither model has a result on the other 28 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index130
Reasoning benchmarks for Kimi K2.5 and o4 Mini
BenchmarkKimi K2.5o4 MiniLeader
GPQA Diamond84.1%74.5%Kimi K2.5
MMMU Pro (Vals AI)84.3%79.7%Kimi K2.5
GPQA Diamond (Artificial Analysis run)87.9%78.4%Kimi K2.5
Humanity's Last Exam (Artificial Analysis run)30.7%16.5%Kimi K2.5
GPQA Diamond (Epoch AI run)Not measured79.6%Only o4 Mini measured
ARC-AGI65.3%58.7%Kimi K2.5
ARC-AGI-211.8%6.1%Kimi K2.5
SimpleBench46.8%38.7%Kimi K2.5
Chess puzzlesNot measured22.0%Only o4 Mini measured
Mystery game puzzlesNot measured5.0%Only o4 Mini measured
VPCTNot measured57.5%Only o4 Mini measured
LMArena Text: hard prompts14531351Kimi K2.5
LMArena Text: expert14661343Kimi K2.5
LMArena Text: hard prompts in English14601366Kimi K2.5
LMArena Vision: diagrams12821197Kimi K2.5
LMArena Vision: homework12971234Kimi K2.5
Kagi LLM Benchmark78.5%67.6%Kimi K2.5
Neither model has a result on the other 5 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index30
Math benchmarks for Kimi K2.5 and o4 Mini
BenchmarkKimi K2.5o4 MiniLeader
AIME95.6%83.7%Kimi K2.5
MATH 500 (Vals AI)Not measured94.2%Only o4 Mini measured
MGSM (Vals AI)Not measured93.4%Only o4 Mini measured
AA Math IndexNot measured90.7Only o4 Mini measured
AIME 2025 (Artificial Analysis run)Not measured90.7%Only o4 Mini measured
OTIS Mock AIMENot measured81.7%Only o4 Mini measured
FrontierMath, tiers 1 to 3Not measured36.1%Only o4 Mini measured
FrontierMath, tier 4Not measured4.9%Only o4 Mini measured
MATH level 5Not measured97.8%Only o4 Mini measured
LMArena Text: math14701389Kimi K2.5
LMArena Text: mathematical work14731379Kimi K2.5
Neither model has a result on the other 2 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index92
Knowledge benchmarks for Kimi K2.5 and o4 Mini
BenchmarkKimi K2.5o4 MiniLeader
MMLU Pro85.9%80.6%Kimi K2.5
LegalBench (Vals AI)Not measured79.2%Only o4 Mini measured
CaseLaw (Vals AI)58.7%Not measuredOnly Kimi K2.5 measured
CorpFin (Vals AI)68.3%59.0%Kimi K2.5
TaxEval (Vals AI)74.2%74.8%o4 Mini
MortgageTax (Vals AI)66.5%64.8%Kimi K2.5
MedQA (Vals AI)94.4%96.0%o4 Mini
MedCode (Vals AI)39.3%33.8%Kimi K2.5
MMLU Pro (Artificial Analysis run)Not measured83.2%Only o4 Mini measured
SimpleQA Verified34.3%19.6%Kimi K2.5
GeoBenchNot measured64.0%Only o4 Mini measured
LMArena Text: science14591354Kimi K2.5
LMArena Text: medicine and healthcare14481361Kimi K2.5
LMArena Text: legal and government14411350Kimi K2.5
LMArena Text: business and finance14271325Kimi K2.5
FACTS Benchmark SuiteNot measured28.1%Only o4 Mini measured
Neither model has a result on the other 2 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index60
Web design benchmarks for Kimi K2.5 and o4 Mini
BenchmarkKimi K2.5o4 MiniLeader
Design Arena: websites1248985Kimi K2.5
Design Arena: UI components1234983Kimi K2.5
Design Arena: game dev12061003Kimi K2.5
Design Arena: data viz1219991Kimi K2.5
Design Arena: 3D1204854Kimi K2.5
Design Arena: all code categories1236976Kimi K2.5
Design Arena: SVG1149Not measuredOnly Kimi K2.5 measured
Design Arena: ASCII art1166Not measuredOnly Kimi K2.5 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index300
General benchmarks for Kimi K2.5 and o4 Mini
BenchmarkKimi K2.5o4 MiniLeader
SAGE (Vals AI)49.9%41.1%Kimi K2.5
MedScribe (Vals AI)76.4%69.1%Kimi K2.5
IFBench70.2%68.7%Kimi K2.5
Long context reasoning (AA-LCR)78.0%61.0%Kimi K2.5
DTBenchNot measured77.6%Only o4 Mini measured
LMCANot measured26.5%Only o4 Mini measured
Fiction.LiveBench86.1%77.8%Kimi K2.5
Creative writing (Lech Mazur)Not measured75.0%Only o4 Mini measured
CL-bench19.3%Not measuredOnly Kimi K2.5 measured
CL-bench Life13.2%Not measuredOnly Kimi K2.5 measured
LMArena Text14451353Kimi K2.5
LMArena Text: instruction following14311320Kimi K2.5
LMArena Text: creative writing14221294Kimi K2.5
LMArena Vision12691194Kimi K2.5
LMArena Text: English14531367Kimi K2.5
LMArena Text: non-English14331337Kimi K2.5
LMArena Text: Chinese14941354Kimi K2.5
LMArena Text: German14411336Kimi K2.5
LMArena Text: Japanese14231308Kimi K2.5
LMArena Text: Korean14121312Kimi K2.5
LMArena Text: French14541364Kimi K2.5
LMArena Text: Spanish14501346Kimi K2.5
LMArena Text: Russian14351334Kimi K2.5
LMArena Text: Polish14451358Kimi K2.5
LMArena Text: multi-turn14441350Kimi K2.5
LMArena Text: longer queries14451315Kimi K2.5
LMArena Text: writing and language14251316Kimi K2.5
LMArena Text: entertainment and media14171308Kimi K2.5
LMArena Vision: English12741200Kimi K2.5
LMArena Vision: Chinese13281218Kimi K2.5
LMArena Vision: reading text12761197Kimi K2.5
LMArena Vision: humor12841199Kimi K2.5
LMArena Vision: captioning12571204Kimi K2.5
LMArena Vision: recognizing things12851218Kimi K2.5
LMArena Vision: creative writing12691166Kimi K2.5
Neither model has a result on the other 11 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Kimi K2.5 and o4 Mini
BenchmarkKimi K2.5o4 MiniLeader
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

Kimi K2.5$0.0020Fits in one request
o4 Mini$0.0033Fits in one request

Kimi K2.5 costs less

Repository review

50K input + 3K output tokens

Kimi K2.5$0.035Fits in one request
o4 Mini$0.068Fits in one request

Kimi K2.5 costs less

Agent loop

200K input + 10K output tokens

Kimi K2.5$0.135Fits in one request
o4 Mini$0.264Exceeds the context window

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 Kimi K2.5 and o4 Mini
SpecificationMoonshotAI Kimi K2.5OpenAI o4 Mini
MakerMoonshot AIOpenAI
WeightsOpen weightsNo public weights listed
ReleasedJanuary 27, 2026April 16, 2025
Context window262K200K
Input $/Mtok$0.532$1.10
Output $/Mtok$2.85$4.40
Coding index38.6Not benchmarked
Independent measurements behind it30
Cost per solved task$0.24Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Kimi K2.5 better than o4 Mini for coding?

Only Kimi K2.5 has a coding score, 38.6 on the Vibe Coding Index; o4 Mini has none, so the two cannot be ranked on coding. Pick o4 Mini for work measured by WeirdML or TaxEval (Vals AI), where it leads.

Which is cheaper, Kimi K2.5 or o4 Mini?

Kimi K2.5 costs $0.532 per million input tokens and $2.85 per million output tokens; o4 Mini costs $1.10 and $4.40. Cost per solved task is known for Kimi K2.5 only.

Which has the bigger context window, Kimi K2.5 or o4 Mini?

Kimi K2.5, with 262K tokens against 200K.

How many benchmarks is the Kimi K2.5 vs o4 Mini comparison based on?

72 benchmarks have a result for both models as of October 22, 2026. Kimi K2.5's coding index rests on 3 independent benchmarks and o4 Mini's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next