LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

Kimi K2.7 Code vs Qwen3.7 Max

Updated October 21, 2026. 43 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Kimi K2.7 Code Replace Qwen3.7 Max
Kimi K2.7 CodeMoonshot AI
Coding
39.1
Overall
25.8
Output $/M
$3.35
5
Qwen3.7 MaxQwen
Coding
45.1
Overall
29.5
Output $/M
$4.425
37
vs boards won of 43 shared, 1 tied

Decision reading

Pick Qwen3.7 Max for coding: it scores 45.1 on the Vibe Coding Index against 39.1 for Kimi K2.7 Code.

Pick Kimi K2.7 Code when output tokens cost 1.3x less, or when a solved task costs $0.35 against $4.69.

Model AKimi K2.7 CodeMoonshot AI39.1/1001 independent benchmark$0.35 per solved taskModel BQwen3.7 MaxQwen45.1/1001 independent benchmark$4.69 per solved task
Shared results
43
Kimi K2.7 Code only
8
Qwen3.7 Max only
16
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 K2.7 Code 39.1 from 1 benchmark, Qwen3.7 Max 45.1 from 1 benchmark.
Widest category
Knowledge, 19.3 points apart on average over 1 shared benchmark.
Widest single gap
FrontierMath, tier 4: Kimi K2.7 Code 12.2% against Qwen3.7 Max 34.1%, 22.0 points apart on a 0 to 100 scale.
Output price
Kimi K2.7 Code $3.35 against Qwen3.7 Max $4.425 per million output tokens, 1.3x apart.
Context window
Kimi K2.7 Code takes 262K tokens, Qwen3.7 Max 1M.

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 K2.7 CodeBarsQwen3.7 MaxTies
Coding39
Intelligence02
Agentic work19
Reasoning05
Math021 tie
Knowledge01
Web design08
General11
All shared boards5371 tie
CodingLike-for-likeFour of these boards feed the Vibe Coding Index39
Coding benchmarks for Kimi K2.7 Code and Qwen3.7 Max
BenchmarkKimi K2.7 CodeQwen3.7 MaxLeader
LMArena WebDev14741515Qwen3.7 Max
AA Coding Index60.866.0Qwen3.7 Max
LiveCodeBench (Vals AI)82.0%87.1%Qwen3.7 Max
SWE-bench Verified (Vals AI)78.2%68.8%Kimi K2.7 Code
Vibe Code Bench (Vals AI)47.2%47.7%Qwen3.7 Max
Code migration (Vals AI)25.4%13.1%Kimi K2.7 Code
ProgramBench (Vals AI)0.0%Not measuredOnly Kimi K2.7 Code measured
Terminal-Bench 2.1 (Vals AI)67.0%61.0%Kimi K2.7 Code
SkillsBench (Vals AI)50.0%Not measuredOnly Kimi K2.7 Code measured
SciCode47.8%49.5%Qwen3.7 Max
Terminal-Bench Hard (Artificial Analysis run)44.7%50.8%Qwen3.7 Max
FrontierCode30.1%Not measuredOnly Kimi K2.7 Code measured
WeirdML54.1%Not measuredOnly Kimi K2.7 Code measured
ALE-Bench886.21189.4Qwen3.7 Max
LMArena WebDev: React14751514Qwen3.7 Max
LMArena WebDev: HTML14511512Qwen3.7 Max
LMArena WebDev: from an image1513Not measuredOnly Kimi K2.7 Code measured
Neither model has a result on the other 21 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Kimi K2.7 Code and Qwen3.7 Max
BenchmarkKimi K2.7 CodeQwen3.7 MaxLeader
AA Intelligence Index25.829.5Qwen3.7 Max
Epoch Capabilities Index150.1154.0Qwen3.7 Max
Agentic workLike-for-likePublic boards, shown as published, kept out of the index19
Agentic work benchmarks for Kimi K2.7 Code and Qwen3.7 Max
BenchmarkKimi K2.7 CodeQwen3.7 MaxLeader
AA Agentic Index21.022.5Qwen3.7 Max
Design Arena: agents building web apps11761208Qwen3.7 Max
Design Arena: agents building full-stack apps11751191Qwen3.7 Max
Design Arena: agents building mobile apps11431153Qwen3.7 Max
Design Arena: agents building Android apps11431165Qwen3.7 Max
Design Arena: agents building Godot games11471187Qwen3.7 Max
Design Arena: agents building games11301161Qwen3.7 Max
Design Arena: HTML slides12011158Kimi K2.7 Code
Design Arena: PowerPoint slides11541196Qwen3.7 Max
Finance Agent (Vals AI)Not measured47.8%Only Qwen3.7 Max measured
Tax Agent Bench (Vals AI)Not measured15.8%Only Qwen3.7 Max measured
Legal Research Bench (Vals AI)Not measured25.5%Only Qwen3.7 Max measured
Excel Modeling Benchmark (Vals AI)Not measured57.0%Only Qwen3.7 Max measured
τ²-Bench Telecom90.1%94.7%Qwen3.7 Max
APEX-Agents37.6%Not measuredOnly Kimi K2.7 Code measured
Vending-Bench 2$5,083Not measuredOnly Kimi K2.7 Code measured
Neither model has a result on the other 23 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
Reasoning benchmarks for Kimi K2.7 Code and Qwen3.7 Max
BenchmarkKimi K2.7 CodeQwen3.7 MaxLeader
GPQA DiamondNot measured90.2%Only Qwen3.7 Max measured
GPQA Diamond (Artificial Analysis run)89.6%92.3%Qwen3.7 Max
Humanity's Last Exam (Artificial Analysis run)35.0%40.5%Qwen3.7 Max
GPQA Diamond (Epoch AI run)87.9%90.9%Qwen3.7 Max
SimpleBench57.9%70.4%Qwen3.7 Max
Chess puzzles14.0%26.0%Qwen3.7 Max
Mystery game puzzlesNot measured32.0%Only Qwen3.7 Max measured
EBR-benchNot measured9.5%Only Qwen3.7 Max measured
Neither model has a result on the other 13 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for Kimi K2.7 Code and Qwen3.7 Max
BenchmarkKimi K2.7 CodeQwen3.7 MaxLeader
OTIS Mock AIME95.6%95.6%Tie
FrontierMath, tiers 1 to 354.0%64.6%Qwen3.7 Max
FrontierMath, tier 412.2%34.1%Qwen3.7 Max
ProofBenchNot measured26.0%Only Qwen3.7 Max measured
Neither model has a result on the other 9 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index01
Knowledge benchmarks for Kimi K2.7 Code and Qwen3.7 Max
BenchmarkKimi K2.7 CodeQwen3.7 MaxLeader
MMLU ProNot measured89.3%Only Qwen3.7 Max measured
LegalBench (Vals AI)Not measured84.9%Only Qwen3.7 Max measured
CorpFin (Vals AI)Not measured63.7%Only Qwen3.7 Max measured
TaxEval (Vals AI)Not measured75.3%Only Qwen3.7 Max measured
MedCode (Vals AI)Not measured38.8%Only Qwen3.7 Max measured
SimpleQA Verified36.5%55.8%Qwen3.7 Max
Neither model has a result on the other 11 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index08
Web design benchmarks for Kimi K2.7 Code and Qwen3.7 Max
BenchmarkKimi K2.7 CodeQwen3.7 MaxLeader
Design Arena: websites12641274Qwen3.7 Max
Design Arena: UI components12591269Qwen3.7 Max
Design Arena: game dev12201265Qwen3.7 Max
Design Arena: data viz12261275Qwen3.7 Max
Design Arena: 3D12411266Qwen3.7 Max
Design Arena: all code categories12531275Qwen3.7 Max
Design Arena: SVG11761216Qwen3.7 Max
Design Arena: ASCII art11961211Qwen3.7 Max
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index11
General benchmarks for Kimi K2.7 Code and Qwen3.7 Max
BenchmarkKimi K2.7 CodeQwen3.7 MaxLeader
MedScribe (Vals AI)Not measured79.4%Only Qwen3.7 Max measured
IFBench63.1%80.5%Qwen3.7 Max
Long context reasoning (AA-LCR)79.3%79.0%Kimi K2.7 Code
DTBenchNot measured92.3%Only Qwen3.7 Max measured
LMCANot measured44.0%Only Qwen3.7 Max measured
Surface Evolver Bench48.8%Not measuredOnly Kimi K2.7 Code measured
Neither model has a result on the other 40 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Kimi K2.7 Code and Qwen3.7 Max
BenchmarkKimi K2.7 CodeQwen3.7 MaxLeader
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 K2.7 Code$0.0023Fits in one request
Qwen3.7 Max$0.0037Fits in one request

Kimi K2.7 Code costs less

Repository review

50K input + 3K output tokens

Kimi K2.7 Code$0.044Fits in one request
Qwen3.7 Max$0.087Fits in one request

Kimi K2.7 Code costs less

Agent loop

200K input + 10K output tokens

Kimi K2.7 Code$0.168Fits in one request
Qwen3.7 Max$0.339Fits in one request

Kimi K2.7 Code 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 K2.7 Code and Qwen3.7 Max
SpecificationMoonshotAI Kimi K2.7 CodeQwen3.7 Max
MakerMoonshot AIQwen
WeightsOpen weightsNo public weights listed
ReleasedJune 12, 2026May 21, 2026
Context window262K1M
Input $/Mtok$0.6712$1.475
Output $/Mtok$3.35$4.425
Coding index39.145.1
Independent measurements behind it11
Cost per solved task$0.35$4.69
SourceOpenRouter listingOpenRouter listing

Questions

Is Kimi K2.7 Code better than Qwen3.7 Max for coding?

Pick Qwen3.7 Max for coding: it scores 45.1 on the Vibe Coding Index against 39.1 for Kimi K2.7 Code. Pick Kimi K2.7 Code when output tokens cost 1.3x less, or when a solved task costs $0.35 against $4.69.

Which is cheaper, Kimi K2.7 Code or Qwen3.7 Max?

Kimi K2.7 Code costs $0.6712 per million input tokens and $3.35 per million output tokens; Qwen3.7 Max costs $1.475 and $4.425. Per solved SWE-bench Verified task, Kimi K2.7 Code costs $0.35 and Qwen3.7 Max $4.69.

Which has the bigger context window, Kimi K2.7 Code or Qwen3.7 Max?

Qwen3.7 Max, with 1M tokens against 262K.

How many benchmarks is the Kimi K2.7 Code vs Qwen3.7 Max comparison based on?

43 benchmarks have a result for both models as of October 21, 2026. Kimi K2.7 Code's coding index rests on 1 independent benchmark and Qwen3.7 Max's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next