LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GPT-5 Mini vs Kimi K2.5

Updated October 21, 2026. 78 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GPT-5 Mini Replace Kimi K2.5
GPT-5 MiniOpenAI
Coding
59.8
Overall
20.6
Output $/M
$2.00
9
Kimi K2.5Moonshot AI
Coding
58.5
Overall
23.5
Output $/M
$2.50
69
vs boards won of 78 shared

Decision reading

GPT-5 Mini and Kimi K2.5 are level for coding, 59.8 against 58.5 on the Vibe Coding Index, so price and context decide.

Pick Kimi K2.5 when what matters is SWE-bench Verified or SWE-bench Multilingual, where it leads. Pick GPT-5 Mini when output tokens cost 1.3x less, or when a solved task costs $0.08 against $0.24.

Model AGPT-5 MiniOpenAI59.8/1001 independent benchmark$0.08 per solved taskModel BKimi K2.5Moonshot AI58.5/1002 independent benchmarks$0.24 per solved task
Shared results
78
GPT-5 Mini only
20
Kimi K2.5 only
20
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
GPT-5 Mini 59.8 from 1 benchmark, Kimi K2.5 58.5 from 2 benchmarks.
Widest category
Agentic work, 16.2 points apart on average over 2 shared benchmarks.
Widest single gap
SWE-bench Multilingual: GPT-5 Mini 39.7% against Kimi K2.5 67.3%, 27.6 points apart on a 0 to 100 scale.
Output price
GPT-5 Mini $2.00 against Kimi K2.5 $2.50 per million output tokens, 1.3x apart.
Context window
GPT-5 Mini takes 400K 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
CategoryGPT-5 MiniBarsKimi K2.5Ties
Coding28
Intelligence02
Agentic work02
Reasoning011
Math03
Knowledge57
Web design08
General228
All shared boards969
CodingLike-for-likeFour of these boards feed the Vibe Coding Index28
Coding benchmarks for GPT-5 Mini and Kimi K2.5
BenchmarkGPT-5 MiniKimi K2.5Leader
SWE-bench Verified59.8%70.8%Kimi K2.5
SWE-bench Multilingual39.7%67.3%Kimi K2.5
LMArena WebDevNot measured1437Only Kimi K2.5 measured
AA Coding IndexNot measured46.8Only Kimi K2.5 measured
LiveCodeBench (Vals AI)86.6%83.9%GPT-5 Mini
SWE-bench Verified (Vals AI)60.8%70.0%Kimi K2.5
Vibe Code Bench (Vals AI)14.2%17.5%Kimi K2.5
Code migration (Vals AI)Not measured7.0%Only Kimi K2.5 measured
Terminal-Bench 2.1 (Vals AI)Not measured42.0%Only Kimi K2.5 measured
LiveCodeBench (Artificial Analysis run)83.8%Not measuredOnly GPT-5 Mini measured
SciCode39.0%Not measuredOnly GPT-5 Mini measured
Terminal-Bench Hard (Artificial Analysis run)33.3%34.8%Kimi K2.5
Terminal-Bench34.8%43.2%Kimi K2.5
WeirdML52.7%45.6%GPT-5 Mini
ALE-BenchNot measured821.6Only Kimi K2.5 measured
LMArena Text: coding14061483Kimi K2.5
LMArena Text: software and IT13941471Kimi K2.5
LMArena WebDev: ReactNot measured1429Only Kimi K2.5 measured
LMArena WebDev: HTMLNot measured1426Only Kimi K2.5 measured
LMArena WebDev: from an imageNot measured1439Only Kimi K2.5 measured
Neither model has a result on the other 18 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for GPT-5 Mini and Kimi K2.5
BenchmarkGPT-5 MiniKimi K2.5Leader
AA Intelligence Index20.623.5Kimi K2.5
Epoch Capabilities Index145.3148.1Kimi K2.5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index02
Agentic work benchmarks for GPT-5 Mini and Kimi K2.5
BenchmarkGPT-5 MiniKimi K2.5Leader
Design Arena: agents building web appsNot measured1136Only Kimi K2.5 measured
Design Arena: agents building full-stack appsNot measured1114Only Kimi K2.5 measured
Design Arena: agents building mobile appsNot measured1147Only Kimi K2.5 measured
Design Arena: agents building Android appsNot measured1124Only Kimi K2.5 measured
Design Arena: agents building Godot gamesNot measured1219Only Kimi K2.5 measured
Finance Agent (Vals AI)Not measured35.8%Only Kimi K2.5 measured
Legal Research Bench (Vals AI)Not measured15.9%Only Kimi K2.5 measured
Excel Modeling Benchmark (Vals AI)Not measured28.5%Only Kimi K2.5 measured
τ²-Bench Telecom71.1%95.9%Kimi K2.5
OSWorldNot measured63.3%Only Kimi K2.5 measured
Vending-Bench 2-$31$1,198Kimi K2.5
Neither model has a result on the other 28 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index011
Reasoning benchmarks for GPT-5 Mini and Kimi K2.5
BenchmarkGPT-5 MiniKimi K2.5Leader
GPQA Diamond80.3%84.1%Kimi K2.5
MMMU Pro (Vals AI)78.9%84.3%Kimi K2.5
GPQA Diamond (Artificial Analysis run)82.8%87.9%Kimi K2.5
Humanity's Last Exam (Artificial Analysis run)21.5%30.7%Kimi K2.5
GPQA Diamond (Epoch AI run)75.0%Not measuredOnly GPT-5 Mini measured
ARC-AGI54.3%65.3%Kimi K2.5
ARC-AGI-24.4%11.8%Kimi K2.5
SimpleBenchNot measured46.8%Only Kimi K2.5 measured
Chess puzzles24.0%Not measuredOnly GPT-5 Mini measured
Mystery game puzzles10.0%Not measuredOnly GPT-5 Mini measured
VPCT40.2%Not measuredOnly GPT-5 Mini measured
LMArena Text: hard prompts13801453Kimi K2.5
LMArena Text: expert13791466Kimi K2.5
LMArena Text: hard prompts in English13831460Kimi K2.5
LMArena Vision: diagrams12021282Kimi K2.5
LMArena Vision: homework12101297Kimi K2.5
Neither model has a result on the other 5 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index03
Math benchmarks for GPT-5 Mini and Kimi K2.5
BenchmarkGPT-5 MiniKimi K2.5Leader
AIME91.5%95.6%Kimi K2.5
MATH 500 (Vals AI)94.8%Not measuredOnly GPT-5 Mini measured
MGSM (Vals AI)92.6%Not measuredOnly GPT-5 Mini measured
AA Math Index90.7Not measuredOnly GPT-5 Mini measured
AIME 2025 (Artificial Analysis run)90.7%Not measuredOnly GPT-5 Mini measured
OTIS Mock AIME86.7%Not measuredOnly GPT-5 Mini measured
FrontierMath, tiers 1 to 346.7%Not measuredOnly GPT-5 Mini measured
FrontierMath, tier 412.2%Not measuredOnly GPT-5 Mini measured
ProofBench9.0%Not measuredOnly GPT-5 Mini measured
MATH level 597.8%Not measuredOnly GPT-5 Mini measured
LMArena Text: math13781470Kimi K2.5
LMArena Text: mathematical work13801473Kimi K2.5
Neither model has a result on the other 1 board here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index57
Knowledge benchmarks for GPT-5 Mini and Kimi K2.5
BenchmarkGPT-5 MiniKimi K2.5Leader
MMLU Pro82.2%85.9%Kimi K2.5
LegalBench (Vals AI)81.8%Not measuredOnly GPT-5 Mini measured
CaseLaw (Vals AI)68.5%58.7%GPT-5 Mini
CorpFin (Vals AI)60.2%68.3%Kimi K2.5
TaxEval (Vals AI)75.2%74.2%GPT-5 Mini
MortgageTax (Vals AI)66.9%66.5%GPT-5 Mini
MedQA (Vals AI)96.1%94.4%GPT-5 Mini
MedCode (Vals AI)43.0%39.3%GPT-5 Mini
MMLU Pro (Artificial Analysis run)83.7%Not measuredOnly GPT-5 Mini measured
SimpleQA Verified21.6%34.3%Kimi K2.5
LMArena Text: science13711459Kimi K2.5
LMArena Text: medicine and healthcare13801448Kimi K2.5
LMArena Text: legal and government13751441Kimi K2.5
LMArena Text: business and finance13581427Kimi K2.5
Neither model has a result on the other 3 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index08
Web design benchmarks for GPT-5 Mini and Kimi K2.5
BenchmarkGPT-5 MiniKimi K2.5Leader
Design Arena: websites11251249Kimi K2.5
Design Arena: UI components11061236Kimi K2.5
Design Arena: game dev11301207Kimi K2.5
Design Arena: data viz11311219Kimi K2.5
Design Arena: 3D10371205Kimi K2.5
Design Arena: all code categories11151236Kimi K2.5
Design Arena: SVG10911149Kimi K2.5
Design Arena: ASCII art11241165Kimi K2.5
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index228
General benchmarks for GPT-5 Mini and Kimi K2.5
BenchmarkGPT-5 MiniKimi K2.5Leader
SAGE (Vals AI)43.0%49.9%Kimi K2.5
MedScribe (Vals AI)80.6%76.4%GPT-5 Mini
IFBench75.4%70.2%GPT-5 Mini
Long context reasoning (AA-LCR)72.3%78.0%Kimi K2.5
DTBench80.5%Not measuredOnly GPT-5 Mini measured
LMCA34.2%Not measuredOnly GPT-5 Mini measured
Fiction.LiveBench69.4%86.1%Kimi K2.5
Creative writing (Lech Mazur)83.1%Not measuredOnly GPT-5 Mini measured
CL-benchNot measured19.3%Only Kimi K2.5 measured
CL-bench LifeNot measured13.2%Only Kimi K2.5 measured
LMArena Text13731445Kimi K2.5
LMArena Text: instruction following13571431Kimi K2.5
LMArena Text: creative writing13251422Kimi K2.5
LMArena Vision12021269Kimi K2.5
LMArena Text: English13791453Kimi K2.5
LMArena Text: non-English13631433Kimi K2.5
LMArena Text: Chinese13861494Kimi K2.5
LMArena Text: German13661441Kimi K2.5
LMArena Text: Japanese13411423Kimi K2.5
LMArena Text: Korean13081412Kimi K2.5
LMArena Text: French13861454Kimi K2.5
LMArena Text: Spanish13541450Kimi K2.5
LMArena Text: Russian13621435Kimi K2.5
LMArena Text: Polish13781445Kimi K2.5
LMArena Text: multi-turn13631444Kimi K2.5
LMArena Text: longer queries13551445Kimi K2.5
LMArena Text: writing and language13441425Kimi K2.5
LMArena Text: entertainment and media13401417Kimi K2.5
LMArena Vision: English12161274Kimi K2.5
LMArena Vision: Chinese12141328Kimi K2.5
LMArena Vision: reading text11991276Kimi K2.5
LMArena Vision: humor12051284Kimi K2.5
LMArena Vision: captioning12171257Kimi K2.5
LMArena Vision: recognizing things12261285Kimi K2.5
LMArena Vision: creative writing11811269Kimi 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 GPT-5 Mini and Kimi K2.5
BenchmarkGPT-5 MiniKimi 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

GPT-5 Mini$0.0013Fits in one request
Kimi K2.5$0.0018Fits in one request

GPT-5 Mini costs less

Repository review

50K input + 3K output tokens

GPT-5 Mini$0.018Fits in one request
Kimi K2.5$0.033Fits in one request

GPT-5 Mini costs less

Agent loop

200K input + 10K output tokens

GPT-5 Mini$0.070Fits in one request
Kimi K2.5$0.125Fits in one request

GPT-5 Mini 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 GPT-5 Mini and Kimi K2.5
SpecificationOpenAI GPT-5 MiniMoonshotAI Kimi K2.5
MakerOpenAIMoonshot AI
WeightsNo public weights listedOpen weights
ReleasedAugust 7, 2025January 27, 2026
Context window400K262K
Input $/Mtok$0.25$0.50
Output $/Mtok$2.00$2.50
Coding index59.858.5
Independent measurements behind it12
Cost per solved task$0.08$0.24
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5 Mini better than Kimi K2.5 for coding?

GPT-5 Mini and Kimi K2.5 are level for coding, 59.8 against 58.5 on the Vibe Coding Index, so price and context decide. Pick Kimi K2.5 when what matters is SWE-bench Verified or SWE-bench Multilingual, where it leads. Pick GPT-5 Mini when output tokens cost 1.3x less, or when a solved task costs $0.08 against $0.24.

Which is cheaper, GPT-5 Mini or Kimi K2.5?

GPT-5 Mini costs $0.25 per million input tokens and $2.00 per million output tokens; Kimi K2.5 costs $0.50 and $2.50. Per solved SWE-bench Verified task, GPT-5 Mini costs $0.08 and Kimi K2.5 $0.24.

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

GPT-5 Mini, with 400K tokens against 262K.

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

78 benchmarks have a result for both models as of October 21, 2026. GPT-5 Mini's coding index rests on 1 independent benchmark and Kimi K2.5's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next