LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GPT-5 vs Kimi K2.6

Updated October 21, 2026. 82 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GPT-5 Replace Kimi K2.6
GPT-5OpenAI
Coding
72.2
Overall
23.0
Output $/M
$10.00
12
Kimi K2.6Moonshot AI
Coding
44.2
Overall
27.0
Output $/M
$3.41
70
vs boards won of 82 shared

Decision reading

Pick GPT-5 for coding: it scores 72.2 on the Vibe Coding Index against 44.2 for Kimi K2.6.

Pick Kimi K2.6 when output tokens cost 2.9x less, or when a solved task costs $0.35 against $0.39.

Model AGPT-5OpenAI72.2/1003 independent benchmarks$0.39 per solved taskModel BKimi K2.6Moonshot AI44.2/1001 independent benchmark$0.35 per solved task
Shared results
82
GPT-5 only
27
Kimi K2.6 only
27
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 72.2 from 3 benchmarks, Kimi K2.6 44.2 from 1 benchmark.
Widest category
Coding, 10.2 points apart on average over 11 shared benchmarks.
Widest single gap
Design Arena: 3D: GPT-5 1056 against Kimi K2.6 1268, 26.5 points apart on a 0 to 100 scale.
Output price
GPT-5 $10.00 against Kimi K2.6 $3.41 per million output tokens, 2.9x apart.
Context window
GPT-5 takes 400K tokens, Kimi K2.6 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-5BarsKimi K2.6Ties
Coding110
Intelligence02
Agentic work01
Reasoning310
Math15
Knowledge48
Web design17
General227
All shared boards1270
CodingLike-for-likeFour of these boards feed the Vibe Coding Index110
Coding benchmarks for GPT-5 and Kimi K2.6
BenchmarkGPT-5Kimi K2.6Leader
Aider polyglot88.0%Not measuredOnly GPT-5 measured
SWE-bench Verified74.4%Not measuredOnly GPT-5 measured
LMArena WebDev14181509Kimi K2.6
AA Coding Index37.861.8Kimi K2.6
LiveCodeBench (Vals AI)85.9%86.8%Kimi K2.6
SWE-bench Verified (Vals AI)69.0%76.2%Kimi K2.6
Vibe Code Bench (Vals AI)20.1%37.9%Kimi K2.6
Code migration (Vals AI)Not measured27.8%Only Kimi K2.6 measured
ProgramBench (Vals AI)Not measured0.0%Only Kimi K2.6 measured
Terminal-Bench 2.1 (Vals AI)Not measured53.6%Only Kimi K2.6 measured
LiveCodeBench (Artificial Analysis run)84.6%Not measuredOnly GPT-5 measured
SciCodeNot measured51.5%Only Kimi K2.6 measured
Terminal-Bench Hard (Artificial Analysis run)37.9%43.9%Kimi K2.6
Terminal-Bench49.6%Not measuredOnly GPT-5 measured
GSO-Bench6.9%Not measuredOnly GPT-5 measured
WeirdML60.7%55.9%GPT-5
ALE-Bench807.61092.7Kimi K2.6
LMArena Text: coding14361489Kimi K2.6
LMArena Text: software and IT14201484Kimi K2.6
LMArena WebDev: ReactNot measured1505Only Kimi K2.6 measured
LMArena WebDev: HTML14271485Kimi K2.6
LMArena WebDev: from an imageNot measured1521Only Kimi K2.6 measured
Neither model has a result on the other 16 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for GPT-5 and Kimi K2.6
BenchmarkGPT-5Kimi K2.6Leader
AA Intelligence Index23.027.0Kimi K2.6
Epoch Capabilities Index150.0150.9Kimi K2.6
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
Agentic work benchmarks for GPT-5 and Kimi K2.6
BenchmarkGPT-5Kimi K2.6Leader
AA Agentic IndexNot measured20.5Only Kimi K2.6 measured
Design Arena: agents building web appsNot measured1268Only Kimi K2.6 measured
Design Arena: agents building full-stack appsNot measured1155Only Kimi K2.6 measured
Design Arena: agents building mobile appsNot measured1156Only Kimi K2.6 measured
Design Arena: agents building Android appsNot measured1166Only Kimi K2.6 measured
Design Arena: agents building Godot gamesNot measured1120Only Kimi K2.6 measured
Design Arena: agents building gamesNot measured1141Only Kimi K2.6 measured
Design Arena: HTML slidesNot measured1184Only Kimi K2.6 measured
Design Arena: PowerPoint slidesNot measured1128Only Kimi K2.6 measured
Design Arena: spreadsheetsNot measured1276Only Kimi K2.6 measured
Finance Agent (Vals AI)Not measured44.9%Only Kimi K2.6 measured
Tax Agent Bench (Vals AI)Not measured12.4%Only Kimi K2.6 measured
Legal Research Bench (Vals AI)Not measured15.9%Only Kimi K2.6 measured
Excel Modeling Benchmark (Vals AI)Not measured57.9%Only Kimi K2.6 measured
τ²-Bench Telecom86.5%95.9%Kimi K2.6
METR time horizons69.6%Not measuredOnly GPT-5 measured
Remote Labor Index1.7%Not measuredOnly GPT-5 measured
GDPval34.8%Not measuredOnly GPT-5 measured
OSWorld 2.0Not measured4.6%Only Kimi K2.6 measured
BALROG32.8%Not measuredOnly GPT-5 measured
DeepResearch Bench49.6%Not measuredOnly GPT-5 measured
Vending-Bench 2Not measured$6,205Only Kimi K2.6 measured
Neither model has a result on the other 17 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index310
Reasoning benchmarks for GPT-5 and Kimi K2.6
BenchmarkGPT-5Kimi K2.6Leader
GPQA Diamond85.6%89.1%Kimi K2.6
MMMU Pro (Vals AI)81.5%86.3%Kimi K2.6
GPQA Diamond (Artificial Analysis run)85.4%91.1%Kimi K2.6
Humanity's Last Exam (Artificial Analysis run)28.5%37.5%Kimi K2.6
GPQA Diamond (Epoch AI run)86.2%90.8%Kimi K2.6
ARC-AGI65.7%Not measuredOnly GPT-5 measured
ARC-AGI-29.9%Not measuredOnly GPT-5 measured
SimpleBench56.7%Not measuredOnly GPT-5 measured
Chess puzzles36.0%20.0%GPT-5
Mystery game puzzles23.0%18.0%GPT-5
EBR-bench12.7%2.4%GPT-5
VPCT66.0%Not measuredOnly GPT-5 measured
LMArena Text: hard prompts14151470Kimi K2.6
LMArena Text: expert14191491Kimi K2.6
LMArena Text: hard prompts in English14191471Kimi K2.6
LMArena Vision: diagrams12521298Kimi K2.6
LMArena Vision: homework12781310Kimi K2.6
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index15
Math benchmarks for GPT-5 and Kimi K2.6
BenchmarkGPT-5Kimi K2.6Leader
AIME93.4%Not measuredOnly GPT-5 measured
MATH 500 (Vals AI)96.0%Not measuredOnly GPT-5 measured
MGSM (Vals AI)92.8%Not measuredOnly GPT-5 measured
AA Math Index94.3Not measuredOnly GPT-5 measured
AIME 2025 (Artificial Analysis run)94.3%Not measuredOnly GPT-5 measured
OTIS Mock AIME91.4%96.1%Kimi K2.6
FrontierMath, tiers 1 to 355.4%57.2%Kimi K2.6
FrontierMath, tier 422.0%25.6%Kimi K2.6
ProofBench18.0%16.0%GPT-5
MATH level 598.1%Not measuredOnly GPT-5 measured
LMArena Text: math14071474Kimi K2.6
LMArena Text: mathematical work14051480Kimi K2.6
Neither model has a result on the other 1 board here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index48
Knowledge benchmarks for GPT-5 and Kimi K2.6
BenchmarkGPT-5Kimi K2.6Leader
MMLU Pro86.5%87.6%Kimi K2.6
LegalBench (Vals AI)86.0%84.7%GPT-5
CaseLaw (Vals AI)66.5%61.2%GPT-5
CorpFin (Vals AI)61.1%66.7%Kimi K2.6
TaxEval (Vals AI)73.4%74.7%Kimi K2.6
MortgageTax (Vals AI)65.5%65.8%Kimi K2.6
MedQA (Vals AI)96.3%Not measuredOnly GPT-5 measured
MedCode (Vals AI)49.6%40.1%GPT-5
MMLU Pro (Artificial Analysis run)87.1%Not measuredOnly GPT-5 measured
SimpleQA Verified50.1%34.9%GPT-5
GeoBench81.0%Not measuredOnly GPT-5 measured
LMArena Text: science14161468Kimi K2.6
LMArena Text: medicine and healthcare14131457Kimi K2.6
LMArena Text: legal and government14311462Kimi K2.6
LMArena Text: business and finance14001445Kimi K2.6
Neither model has a result on the other 2 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index17
Web design benchmarks for GPT-5 and Kimi K2.6
BenchmarkGPT-5Kimi K2.6Leader
Design Arena: websites11851269Kimi K2.6
Design Arena: UI components11791266Kimi K2.6
Design Arena: game dev11851251Kimi K2.6
Design Arena: data viz12211249Kimi K2.6
Design Arena: 3D10561268Kimi K2.6
Design Arena: all code categories11681272Kimi K2.6
Design Arena: SVG11881181GPT-5
Design Arena: ASCII art11401164Kimi K2.6
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index227
General benchmarks for GPT-5 and Kimi K2.6
BenchmarkGPT-5Kimi K2.6Leader
Public Benefits Bench (Vals AI)Not measured56.6%Only Kimi K2.6 measured
SAGE (Vals AI)43.7%50.2%Kimi K2.6
MedScribe (Vals AI)83.7%78.2%GPT-5
Vals Multimodal IndexNot measured56.4%Only Kimi K2.6 measured
IFBench73.1%76.0%Kimi K2.6
Long context reasoning (AA-LCR)78.2%81.0%Kimi K2.6
DTBench90.7%90.9%Kimi K2.6
LMCA40.0%37.3%GPT-5
Fiction.LiveBench97.2%Not measuredOnly GPT-5 measured
Creative writing (Lech Mazur)86.0%Not measuredOnly GPT-5 measured
Furniture assemblyNot measured21.7%Only Kimi K2.6 measured
ExploitBenchNot measured18.4%Only Kimi K2.6 measured
LMArena Text14061455Kimi K2.6
LMArena Text: instruction following13881451Kimi K2.6
LMArena Text: creative writing13651434Kimi K2.6
LMArena Vision12321282Kimi K2.6
LMArena Text: English14091460Kimi K2.6
LMArena Text: non-English13971446Kimi K2.6
LMArena Text: Chinese14221520Kimi K2.6
LMArena Text: German14161451Kimi K2.6
LMArena Text: Japanese14091443Kimi K2.6
LMArena Text: Korean13601427Kimi K2.6
LMArena Text: French14091471Kimi K2.6
LMArena Text: Spanish13991464Kimi K2.6
LMArena Text: Russian14071446Kimi K2.6
LMArena Text: Polish14311438Kimi K2.6
LMArena Text: multi-turn14261453Kimi K2.6
LMArena Text: longer queries13991468Kimi K2.6
LMArena Text: writing and language13841439Kimi K2.6
LMArena Text: entertainment and media13791428Kimi K2.6
LMArena Vision: English12311287Kimi K2.6
LMArena Vision: Chinese12671346Kimi K2.6
LMArena Vision: reading text12461291Kimi K2.6
LMArena Vision: humor12191280Kimi K2.6
LMArena Vision: captioning1212Not measuredOnly GPT-5 measured
LMArena Vision: recognizing things1263Not measuredOnly GPT-5 measured
LMArena Vision: creative writing12101273Kimi K2.6
LMArena DocumentNot measured1451Only Kimi K2.6 measured
Neither model has a result on the other 8 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-5 and Kimi K2.6
BenchmarkGPT-5Kimi K2.6Leader
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$0.0063Fits in one request
Kimi K2.6$0.0024Fits in one request

Kimi K2.6 costs less

Repository review

50K input + 3K output tokens

GPT-5$0.092Fits in one request
Kimi K2.6$0.043Fits in one request

Kimi K2.6 costs less

Agent loop

200K input + 10K output tokens

GPT-5$0.350Fits in one request
Kimi K2.6$0.164Fits in one request

Kimi K2.6 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 and Kimi K2.6
SpecificationOpenAI GPT-5MoonshotAI Kimi K2.6
MakerOpenAIMoonshot AI
WeightsNo public weights listedOpen weights
ReleasedAugust 7, 2025April 20, 2026
Context window400K262K
Input $/Mtok$1.25$0.65
Output $/Mtok$10.00$3.41
Coding index72.244.2
Independent measurements behind it31
Cost per solved task$0.39$0.35
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5 better than Kimi K2.6 for coding?

Pick GPT-5 for coding: it scores 72.2 on the Vibe Coding Index against 44.2 for Kimi K2.6. Pick Kimi K2.6 when output tokens cost 2.9x less, or when a solved task costs $0.35 against $0.39.

Which is cheaper, GPT-5 or Kimi K2.6?

GPT-5 costs $1.25 per million input tokens and $10.00 per million output tokens; Kimi K2.6 costs $0.65 and $3.41. Per solved SWE-bench Verified task, GPT-5 costs $0.39 and Kimi K2.6 $0.35.

Which has the bigger context window, GPT-5 or Kimi K2.6?

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

How many benchmarks is the GPT-5 vs Kimi K2.6 comparison based on?

82 benchmarks have a result for both models as of October 21, 2026. GPT-5's coding index rests on 3 independent benchmarks and Kimi K2.6's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next