LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

Kimi K2.5 vs o3

Updated October 21, 2026. 73 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Kimi K2.5 Replace o3
Kimi K2.5Moonshot AI
Coding
58.5
Overall
23.5
Output $/M
$2.50
60
o3OpenAI
Coding
70.7
Overall
20.2
Output $/M
$8.00
12
vs boards won of 73 shared, 1 tied

Decision reading

Pick o3 for coding: it scores 70.7 on the Vibe Coding Index against 58.5 for Kimi K2.5.

Pick Kimi K2.5 when output tokens cost 3.2x less, or when a solved task costs $0.24 against $0.571.

Model AKimi K2.5Moonshot AI58.5/1002 independent benchmarks$0.24 per solved taskModel Bo3OpenAI70.7/1002 independent benchmarks$0.571 per solved task
Shared results
73
Kimi K2.5 only
25
o3 only
24
Like-for-like categories
8 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

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 58.5 from 2 benchmarks, o3 70.7 from 2 benchmarks.
Widest category
Agentic work, 27.8 points apart on average over 2 shared benchmarks.
Widest single gap
OSWorld: Kimi K2.5 63.3% against o3 23.0%, 40.3 points apart on a 0 to 100 scale.
Output price
Kimi K2.5 $2.50 against o3 $8.00 per million output tokens, 3.2x apart.
Context window
Kimi K2.5 takes 262K tokens, o3 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 K2.5Barso3Ties
Coding34
Intelligence20
Agentic work20
Reasoning1011 tie
Math30
Knowledge74
Web design50
General283
All shared boards60121 tie
CodingLike-for-likeFour of these boards feed the Vibe Coding Index34
Coding benchmarks for Kimi K2.5 and o3
BenchmarkKimi K2.5o3Leader
Aider polyglotNot measured81.3%Only o3 measured
SWE-bench Verified70.8%58.4%Kimi K2.5
SWE-bench MultimodalNot measured36.0%Only o3 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%83.9%o3
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 measured80.8%Only o3 measured
Terminal-Bench Hard (Artificial Analysis run)34.8%37.1%o3
Terminal-Bench43.2%Not measuredOnly Kimi K2.5 measured
GSO-BenchNot measured8.8%Only o3 measured
WeirdML45.6%52.4%o3
ALE-Bench821.6933.5o3
LMArena Text: coding14831407Kimi K2.5
LMArena Text: software and IT14711415Kimi 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
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 o3
BenchmarkKimi K2.5o3Leader
AA Intelligence Index23.520.2Kimi K2.5
Epoch Capabilities Index148.1146.7Kimi K2.5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index20
Agentic work benchmarks for Kimi K2.5 and o3
BenchmarkKimi K2.5o3Leader
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 apps1124Not 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%80.7%Kimi K2.5
METR time horizonsNot measured65.4%Only o3 measured
GDPvalNot measured30.8%Only o3 measured
DeepResearch BenchNot measured45.2%Only o3 measured
OSWorld63.3%23.0%Kimi K2.5
Vending-Bench 2$1,198Not measuredOnly Kimi K2.5 measured
Neither model has a result on the other 25 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index101
Reasoning benchmarks for Kimi K2.5 and o3
BenchmarkKimi K2.5o3Leader
GPQA Diamond84.1%84.1%Tie
MMMU Pro (Vals AI)84.3%80.4%Kimi K2.5
GPQA Diamond (Artificial Analysis run)87.9%82.7%Kimi K2.5
Humanity's Last Exam (Artificial Analysis run)30.7%20.1%Kimi K2.5
GPQA Diamond (Epoch AI run)Not measured81.8%Only o3 measured
ARC-AGI65.3%60.8%Kimi K2.5
ARC-AGI-211.8%6.5%Kimi K2.5
SimpleBench46.8%53.1%o3
Chess puzzlesNot measured37.0%Only o3 measured
Mystery game puzzlesNot measured29.0%Only o3 measured
VPCTNot measured52.0%Only o3 measured
LMArena Text: hard prompts14531402Kimi K2.5
LMArena Text: expert14661402Kimi K2.5
LMArena Text: hard prompts in English14601411Kimi K2.5
LMArena Vision: diagrams12821220Kimi K2.5
LMArena Vision: homework12971234Kimi 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 o3
BenchmarkKimi K2.5o3Leader
AIME95.6%85.3%Kimi K2.5
MATH 500 (Vals AI)Not measured94.6%Only o3 measured
MGSM (Vals AI)Not measured91.8%Only o3 measured
AA Math IndexNot measured88.3Only o3 measured
AIME 2025 (Artificial Analysis run)Not measured88.3%Only o3 measured
OTIS Mock AIMENot measured84.4%Only o3 measured
FrontierMath, tiers 1 to 3Not measured33.3%Only o3 measured
MATH level 5Not measured97.8%Only o3 measured
LMArena Text: math14701426Kimi K2.5
LMArena Text: mathematical work14731423Kimi K2.5
Neither model has a result on the other 3 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index74
Knowledge benchmarks for Kimi K2.5 and o3
BenchmarkKimi K2.5o3Leader
MMLU Pro85.9%85.6%Kimi K2.5
LegalBench (Vals AI)Not measured83.8%Only o3 measured
CaseLaw (Vals AI)58.7%Not measuredOnly Kimi K2.5 measured
CorpFin (Vals AI)68.3%59.7%Kimi K2.5
TaxEval (Vals AI)74.2%74.6%o3
MortgageTax (Vals AI)66.5%65.7%Kimi K2.5
MedQA (Vals AI)94.4%96.1%o3
MedCode (Vals AI)39.3%47.3%o3
MMLU Pro (Artificial Analysis run)Not measured85.3%Only o3 measured
SimpleQA Verified34.3%49.4%o3
GeoBenchNot measured74.0%Only o3 measured
LMArena Text: science14591429Kimi K2.5
LMArena Text: medicine and healthcare14481448Kimi K2.5
LMArena Text: legal and government14411428Kimi K2.5
LMArena Text: business and finance14271393Kimi K2.5
Neither model has a result on the other 2 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index50
Web design benchmarks for Kimi K2.5 and o3
BenchmarkKimi K2.5o3Leader
Design Arena: websites12491036Kimi K2.5
Design Arena: UI components12361016Kimi K2.5
Design Arena: game dev12071034Kimi K2.5
Design Arena: data viz12191200Kimi K2.5
Design Arena: 3D1205Not measuredOnly Kimi K2.5 measured
Design Arena: all code categories12361021Kimi K2.5
Design Arena: SVG1149Not measuredOnly Kimi K2.5 measured
Design Arena: ASCII art1165Not 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 index283
General benchmarks for Kimi K2.5 and o3
BenchmarkKimi K2.5o3Leader
SAGE (Vals AI)49.9%41.8%Kimi K2.5
MedScribe (Vals AI)76.4%76.7%o3
IFBench70.2%71.4%o3
Long context reasoning (AA-LCR)78.0%74.7%Kimi K2.5
DTBenchNot measured84.8%Only o3 measured
LMCANot measured39.7%Only o3 measured
Fiction.LiveBench86.1%88.9%o3
Creative writing (Lech Mazur)Not measured83.9%Only o3 measured
CL-bench19.3%17.8%Kimi K2.5
CL-bench Life13.2%Not measuredOnly Kimi K2.5 measured
LMArena Text14451410Kimi K2.5
LMArena Text: instruction following14311368Kimi K2.5
LMArena Text: creative writing14221359Kimi K2.5
LMArena Vision12691214Kimi K2.5
LMArena Text: English14531417Kimi K2.5
LMArena Text: non-English14331401Kimi K2.5
LMArena Text: Chinese14941437Kimi K2.5
LMArena Text: German14411420Kimi K2.5
LMArena Text: Japanese14231403Kimi K2.5
LMArena Text: Korean14121370Kimi K2.5
LMArena Text: French14541430Kimi K2.5
LMArena Text: Spanish14501395Kimi K2.5
LMArena Text: Russian14351406Kimi K2.5
LMArena Text: Polish14451438Kimi K2.5
LMArena Text: multi-turn14441405Kimi K2.5
LMArena Text: longer queries14451372Kimi K2.5
LMArena Text: writing and language14251372Kimi K2.5
LMArena Text: entertainment and media14171373Kimi K2.5
LMArena Vision: English12741221Kimi K2.5
LMArena Vision: Chinese13281237Kimi K2.5
LMArena Vision: reading text12761218Kimi K2.5
LMArena Vision: humor12841222Kimi K2.5
LMArena Vision: captioning12571210Kimi K2.5
LMArena Vision: recognizing things12851234Kimi K2.5
LMArena Vision: creative writing12691201Kimi 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 o3
BenchmarkKimi K2.5o3Leader
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$0.0018Fits in one request
o3$0.0060Fits in one request

Kimi K2.5 costs less

Repository review

50K input + 3K output tokens

Kimi K2.5$0.033Fits in one request
o3$0.124Fits in one request

Kimi K2.5 costs less

Agent loop

200K input + 10K output tokens

Kimi K2.5$0.125Fits in one request
o3$0.480Exceeds 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 o3
SpecificationMoonshotAI Kimi K2.5OpenAI o3
MakerMoonshot AIOpenAI
WeightsOpen weightsNo public weights listed
ReleasedJanuary 27, 2026April 16, 2025
Context window262K200K
Input $/Mtok$0.50$2.00
Output $/Mtok$2.50$8.00
Coding index58.570.7
Independent measurements behind it22
Cost per solved task$0.24$0.571
SourceOpenRouter listingOpenRouter listing

Questions

Is Kimi K2.5 better than o3 for coding?

Pick o3 for coding: it scores 70.7 on the Vibe Coding Index against 58.5 for Kimi K2.5. Pick Kimi K2.5 when output tokens cost 3.2x less, or when a solved task costs $0.24 against $0.571.

Which is cheaper, Kimi K2.5 or o3?

Kimi K2.5 costs $0.50 per million input tokens and $2.50 per million output tokens; o3 costs $2.00 and $8.00. Per solved SWE-bench Verified task, Kimi K2.5 costs $0.24 and o3 $0.571.

Which has the bigger context window, Kimi K2.5 or o3?

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

How many benchmarks is the Kimi K2.5 vs o3 comparison based on?

73 benchmarks have a result for both models as of October 21, 2026. Kimi K2.5's coding index rests on 2 independent benchmarks and o3's on 2. 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/kimi-k2-5-vs-o3" title="Kimi K2.5 vs o3, 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/kimi-k2-5-vs-o3">LLMBenchmark.ai, Kimi K2.5 vs o3</a></p>