Enter opens the highlighted result. Esc closes.

Qwen3.6 Plus vs Qwen3.7 Max

Qwen3.7 Max leads Qwen3.6 Plus on coding, 54.1 to 44.1 on the Vibe Coding Index, and Qwen3.6 Plus costs 2.3x less per output token ($1.95 vs $4.425 per million). Updated October 21, 2026; 45 benchmarks have a result for both.

Replace Qwen3.6 Plus Replace Qwen3.7 Max
Qwen3.6 PlusQwen
Coding
44.1
Overall
56.2
Output $/M
$1.95
2
Qwen3.7 MaxQwen
Coding
54.1
Overall
61.7
Output $/M
$4.425
43
vs boards won of 45 shared

Qwen3.6 Plus or Qwen3.7 Max: which to pick

Pick Qwen3.7 Max for coding: it scores 54.1 on the Vibe Coding Index against 44.1 for Qwen3.6 Plus.

Pick Qwen3.6 Plus for output tokens 2.3x cheaper ($1.95 vs $4.425 per million) and a lower cost per solved task ($1.19 against $4.69).

Model AQwen3.6 PlusQwen44.1/1003 independent benchmarks$1.19 per solved taskModel BQwen3.7 MaxQwen54.1/1003 independent benchmarks$4.69 per solved task
Shared results
45
Qwen3.6 Plus only
38
Qwen3.7 Max only
14
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
Qwen3.6 Plus 44.1 from 3 benchmarks, Qwen3.7 Max 54.1 from 3 benchmarks.
Widest category
Math, 14.3 points apart on average over 2 shared benchmarks.
Widest single gap
FrontierMath, tiers 1 to 3: Qwen3.6 Plus 38.2% against Qwen3.7 Max 64.6%, 26.3 points apart on a 0 to 100 scale.
Output price
Qwen3.6 Plus $1.95 against Qwen3.7 Max $4.425 per million output tokens, 2.3x apart.
Context window
Both take 1M tokens.

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
CategoryQwen3.6 PlusBarsQwen3.7 MaxTies
Coding110
Intelligence02
Agentic work14
Reasoning06
Math02
Knowledge06
Web design08
General05
All shared boards243
CodingLike-for-likeSix of these boards feed the Vibe Coding Index110
Coding benchmarks for Qwen3.6 Plus and Qwen3.7 Max
BenchmarkQwen3.6 PlusQwen3.7 MaxLeader
LMArena WebDev14611515Qwen3.7 Max
AA Coding Index54.566.0Qwen3.7 Max
LiveCodeBench (Vals AI)86.0%87.1%Qwen3.7 Max
SWE-bench Verified (Vals AI)73.4%68.8%Qwen3.6 Plus
Vibe Code Bench (Vals AI)25.6%47.7%Qwen3.7 Max
Code migration (Vals AI)11.1%13.1%Qwen3.7 Max
ProgramBench (Vals AI)0.0%Not measuredOnly Qwen3.6 Plus measured
Terminal-Bench 2.1 (Vals AI)53.2%61.0%Qwen3.7 Max
SciCodeNot measured49.5%Only Qwen3.7 Max measured
Terminal-Bench Hard (Artificial Analysis run)43.9%50.8%Qwen3.7 Max
ALE-Bench670.11189.4Qwen3.7 Max
LMArena Text: coding1467Not measuredOnly Qwen3.6 Plus measured
LMArena Text: software and IT1462Not measuredOnly Qwen3.6 Plus measured
LMArena WebDev: React14541514Qwen3.7 Max
LMArena WebDev: HTML14631512Qwen3.7 Max
LMArena WebDev: from an image1472Not measuredOnly Qwen3.6 Plus measured
Neither model has a result on the other 23 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Qwen3.6 Plus and Qwen3.7 Max
BenchmarkQwen3.6 PlusQwen3.7 MaxLeader
AA Intelligence Index27.029.5Qwen3.7 Max
Epoch Capabilities Index147.3154.0Qwen3.7 Max
Agentic workLike-for-likePublic boards, shown as published, kept out of the index14
Agentic work benchmarks for Qwen3.6 Plus and Qwen3.7 Max
BenchmarkQwen3.6 PlusQwen3.7 MaxLeader
AA Agentic IndexNot measured22.5Only Qwen3.7 Max measured
Design Arena: agents building web appsNot measured1208Only Qwen3.7 Max measured
Design Arena: agents building full-stack appsNot measured1191Only Qwen3.7 Max measured
Design Arena: agents building mobile appsNot measured1155Only Qwen3.7 Max measured
Design Arena: agents building Android appsNot measured1159Only Qwen3.7 Max measured
Design Arena: agents building Godot gamesNot measured1187Only Qwen3.7 Max measured
Design Arena: agents building gamesNot measured1161Only Qwen3.7 Max measured
Design Arena: HTML slidesNot measured1158Only Qwen3.7 Max measured
Design Arena: PowerPoint slidesNot measured1197Only Qwen3.7 Max measured
Finance Agent (Vals AI)40.9%47.8%Qwen3.7 Max
Tax Agent Bench (Vals AI)7.7%15.8%Qwen3.7 Max
Legal Research Bench (Vals AI)14.9%25.5%Qwen3.7 Max
Excel Modeling Benchmark (Vals AI)32.9%57.0%Qwen3.7 Max
τ²-Bench Telecom97.7%94.7%Qwen3.6 Plus
Vending-Bench 2$5,115Not measuredOnly Qwen3.6 Plus measured
Neither model has a result on the other 26 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index06
Reasoning benchmarks for Qwen3.6 Plus and Qwen3.7 Max
BenchmarkQwen3.6 PlusQwen3.7 MaxLeader
GPQA Diamond87.4%90.2%Qwen3.7 Max
MMMU Pro (Vals AI)84.2%Not measuredOnly Qwen3.6 Plus measured
GPQA Diamond (Artificial Analysis run)88.2%92.3%Qwen3.7 Max
Humanity's Last Exam (Artificial Analysis run)27.8%40.5%Qwen3.7 Max
GPQA Diamond (Epoch AI run)88.4%90.9%Qwen3.7 Max
SimpleBenchNot measured70.4%Only Qwen3.7 Max measured
Chess puzzles14.0%26.0%Qwen3.7 Max
Mystery game puzzles12.0%32.0%Qwen3.7 Max
EBR-benchNot measured9.5%Only Qwen3.7 Max measured
LMArena Text: hard prompts1448Not measuredOnly Qwen3.6 Plus measured
LMArena Text: expert1454Not measuredOnly Qwen3.6 Plus measured
LMArena Text: hard prompts in English1455Not measuredOnly Qwen3.6 Plus measured
Neither model has a result on the other 10 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for Qwen3.6 Plus and Qwen3.7 Max
BenchmarkQwen3.6 PlusQwen3.7 MaxLeader
AIME94.6%Not measuredOnly Qwen3.6 Plus measured
OTIS Mock AIME93.3%95.6%Qwen3.7 Max
FrontierMath, tiers 1 to 338.2%64.6%Qwen3.7 Max
FrontierMath, tier 4Not measured34.1%Only Qwen3.7 Max measured
ProofBenchNot measured26.0%Only Qwen3.7 Max measured
LMArena Text: math1449Not measuredOnly Qwen3.6 Plus measured
LMArena Text: mathematical work1448Not measuredOnly Qwen3.6 Plus measured
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index06
Knowledge benchmarks for Qwen3.6 Plus and Qwen3.7 Max
BenchmarkQwen3.6 PlusQwen3.7 MaxLeader
MMLU Pro87.7%89.3%Qwen3.7 Max
LegalBench (Vals AI)84.2%84.9%Qwen3.7 Max
CaseLaw (Vals AI)51.5%Not measuredOnly Qwen3.6 Plus measured
CorpFin (Vals AI)61.9%63.7%Qwen3.7 Max
TaxEval (Vals AI)74.7%75.3%Qwen3.7 Max
MortgageTax (Vals AI)68.0%Not measuredOnly Qwen3.6 Plus measured
MedCode (Vals AI)36.9%38.8%Qwen3.7 Max
SimpleQA Verified44.1%55.8%Qwen3.7 Max
LMArena Text: science1439Not measuredOnly Qwen3.6 Plus measured
LMArena Text: medicine and healthcare1442Not measuredOnly Qwen3.6 Plus measured
LMArena Text: legal and government1437Not measuredOnly Qwen3.6 Plus measured
LMArena Text: business and finance1430Not measuredOnly Qwen3.6 Plus measured
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index08
Web design benchmarks for Qwen3.6 Plus and Qwen3.7 Max
BenchmarkQwen3.6 PlusQwen3.7 MaxLeader
Design Arena: websites12431274Qwen3.7 Max
Design Arena: UI components12331270Qwen3.7 Max
Design Arena: game dev12111265Qwen3.7 Max
Design Arena: data viz12281275Qwen3.7 Max
Design Arena: 3D12021266Qwen3.7 Max
Design Arena: all code categories12361275Qwen3.7 Max
Design Arena: SVG11611217Qwen3.7 Max
Design Arena: ASCII art11131211Qwen3.7 Max
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index05
General benchmarks for Qwen3.6 Plus and Qwen3.7 Max
BenchmarkQwen3.6 PlusQwen3.7 MaxLeader
SAGE (Vals AI)44.9%Not measuredOnly Qwen3.6 Plus measured
MedScribe (Vals AI)77.0%79.4%Qwen3.7 Max
Vals Multimodal Index51.5%Not measuredOnly Qwen3.6 Plus measured
IFBench75.2%80.5%Qwen3.7 Max
Long context reasoning (AA-LCR)78.3%79.0%Qwen3.7 Max
DTBench81.9%92.3%Qwen3.7 Max
LMCA33.1%44.0%Qwen3.7 Max
CL-bench20.3%Not measuredOnly Qwen3.6 Plus measured
LMArena Text1436Not measuredOnly Qwen3.6 Plus measured
LMArena Text: instruction following1424Not measuredOnly Qwen3.6 Plus measured
LMArena Text: creative writing1404Not measuredOnly Qwen3.6 Plus measured
LMArena Text: English1445Not measuredOnly Qwen3.6 Plus measured
LMArena Text: non-English1424Not measuredOnly Qwen3.6 Plus measured
LMArena Text: Chinese1477Not measuredOnly Qwen3.6 Plus measured
LMArena Text: German1451Not measuredOnly Qwen3.6 Plus measured
LMArena Text: Japanese1390Not measuredOnly Qwen3.6 Plus measured
LMArena Text: Korean1379Not measuredOnly Qwen3.6 Plus measured
LMArena Text: French1455Not measuredOnly Qwen3.6 Plus measured
LMArena Text: Spanish1432Not measuredOnly Qwen3.6 Plus measured
LMArena Text: Russian1434Not measuredOnly Qwen3.6 Plus measured
LMArena Text: Polish1433Not measuredOnly Qwen3.6 Plus measured
LMArena Text: multi-turn1438Not measuredOnly Qwen3.6 Plus measured
LMArena Text: longer queries1439Not measuredOnly Qwen3.6 Plus measured
LMArena Text: writing and language1418Not measuredOnly Qwen3.6 Plus measured
LMArena Text: entertainment and media1402Not measuredOnly Qwen3.6 Plus measured
Neither model has a result on the other 21 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Qwen3.6 Plus and Qwen3.7 Max
BenchmarkQwen3.6 PlusQwen3.7 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

Qwen3.6 Plus$0.0013Fits in one request
Qwen3.7 Max$0.0037Fits in one request

Qwen3.6 Plus costs less

Repository review

50K input + 3K output tokens

Qwen3.6 Plus$0.022Fits in one request
Qwen3.7 Max$0.087Fits in one request

Qwen3.6 Plus costs less

Agent loop

200K input + 10K output tokens

Qwen3.6 Plus$0.085Fits in one request
Qwen3.7 Max$0.339Fits in one request

Qwen3.6 Plus 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 Qwen3.6 Plus and Qwen3.7 Max
SpecificationQwen3.6 PlusQwen3.7 Max
MakerQwenQwen
WeightsNo public weights listedNo public weights listed
ReleasedApril 2, 2026May 21, 2026
Context window1M1M
Input $/Mtok$0.325$1.475
Output $/Mtok$1.95$4.425
Coding index44.154.1
Independent measurements behind it33
Cost per solved task$1.19$4.69
SourceOpenRouter listingOpenRouter listing

Questions

Is Qwen3.6 Plus better than Qwen3.7 Max for coding?

Pick Qwen3.7 Max for coding: it scores 54.1 on the Vibe Coding Index against 44.1 for Qwen3.6 Plus. Pick Qwen3.6 Plus for output tokens 2.3x cheaper ($1.95 vs $4.425 per million) and a lower cost per solved task ($1.19 against $4.69).

Which is cheaper, Qwen3.6 Plus or Qwen3.7 Max?

Qwen3.6 Plus costs $0.325 per million input tokens and $1.95 per million output tokens; Qwen3.7 Max costs $1.475 and $4.425. Per solved SWE-bench Verified task, Qwen3.6 Plus costs $1.19 and Qwen3.7 Max $4.69.

Which has the bigger context window, Qwen3.6 Plus or Qwen3.7 Max?

Neither: both take 1M tokens.

How many benchmarks is the Qwen3.6 Plus vs Qwen3.7 Max comparison based on?

45 benchmarks have a result for both models as of October 21, 2026. Qwen3.6 Plus's coding index rests on 3 independent benchmarks and Qwen3.7 Max's on 3. 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/qwen3-6-plus-vs-qwen3-7-max" title="Qwen3.6 Plus vs Qwen3.7 Max, 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/qwen3-6-plus-vs-qwen3-7-max">LLMBenchmark.ai, Qwen3.6 Plus vs Qwen3.7 Max</a></p>