Enter opens the highlighted result. Esc closes.

Qwen3.7 Max vs Qwen4 Max

Qwen4 Max leads Qwen3.7 Max on coding, 74.0 to 54.1 on the Vibe Coding Index, at about the same output price ($4.275 vs $4.425 per million). Updated November 22, 2026; 10 benchmarks have a result for both.

Qwen3.7 MaxQwen
Coding
54.1
Overall
67.1
Output $/M
$4.425
0
Qwen4 MaxQwen
Coding
74.0
Overall
66.6
Output $/M
$4.275
10
vs boards won of 10 shared

Qwen3.7 Max or Qwen4 Max: which to pick

Pick Qwen4 Max for coding: it scores 74.0 on the Vibe Coding Index against 54.1 for Qwen3.7 Max.

Qwen3.7 Max is neither cheaper nor larger in context, and leads no benchmark both models were measured on.

Shared results
10
Qwen3.7 Max only
51
Qwen4 Max only
16
Like-for-like categories
3 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
Qwen3.7 Max 54.1 from 3 benchmarks, Qwen4 Max 74.0 from 4 benchmarks.
Widest category
General, 56.2 points apart on average over 1 shared benchmark.
Widest single gap
GDPval-AA: Qwen3.7 Max 1134 against Qwen4 Max 1809, 56.2 points apart on a 0 to 100 scale.
Output price
Qwen3.7 Max $4.425 against Qwen4 Max $4.275 per million output tokens, 1.0x 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
CategoryQwen3.7 MaxBarsQwen4 MaxTies
Coding07
Agentic work02
General01
All shared boards010
CodingLike-for-likeSix of these boards feed the Vibe Coding Index07
Coding benchmarks for Qwen3.7 Max and Qwen4 Max
BenchmarkQwen3.7 MaxQwen4 MaxLeader
LiveBench codingNot measured69.8%Only Qwen4 Max measured
LMArena WebDev15161751Qwen4 Max
LiveCodeBench (Vals AI)87.1%Not measuredOnly Qwen3.7 Max measured
SWE-bench Verified (Vals AI)68.8%Not measuredOnly Qwen3.7 Max measured
Terminal-Bench 4 (Vals AI)Not measured48.5%Only Qwen4 Max measured
Vibe Code Bench (Vals AI)47.7%86.7%Qwen4 Max
Code migration (Vals AI)13.1%Not measuredOnly Qwen3.7 Max measured
Terminal-Bench 2.1 (Vals AI)61.0%Not measuredOnly Qwen3.7 Max measured
SciCode49.5%Not measuredOnly Qwen3.7 Max measured
Terminal-Bench Hard (Artificial Analysis run)50.8%Not measuredOnly Qwen3.7 Max measured
Terminal-Bench 4.0 (Artificial Analysis run)1.5%48.5%Qwen4 Max
FrontierCode 2.0Not measured26.5%Only Qwen4 Max measured
ALE-Bench1189.41831.0Qwen4 Max
LMArena WebDev: React15131785Qwen4 Max
LMArena WebDev: HTML15141752Qwen4 Max
LMArena WebDev: from an imageNot measured1675Only Qwen4 Max measured
GameBench 239.282.4Qwen4 Max
Neither model has a result on the other 35 boards here
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
Intelligence benchmarks for Qwen3.7 Max and Qwen4 Max
BenchmarkQwen3.7 MaxQwen4 MaxLeader
AA Intelligence IndexNot measured42.9Only Qwen4 Max measured
Epoch Capabilities Index154.6Not measuredOnly Qwen3.7 Max measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index02
Agentic work benchmarks for Qwen3.7 Max and Qwen4 Max
BenchmarkQwen3.7 MaxQwen4 MaxLeader
Design Arena: agents building web apps1176Not measuredOnly Qwen3.7 Max measured
Design Arena: agents building full-stack apps1171Not measuredOnly Qwen3.7 Max measured
Design Arena: agents building mobile apps1151Not measuredOnly Qwen3.7 Max measured
Design Arena: agents building Android apps1146Not measuredOnly Qwen3.7 Max measured
Design Arena: agents building Godot games1156Not measuredOnly Qwen3.7 Max measured
Design Arena: agents building games1105Not measuredOnly Qwen3.7 Max measured
Design Arena: HTML slides1058Not measuredOnly Qwen3.7 Max measured
Design Arena: PowerPoint slides1151Not measuredOnly Qwen3.7 Max measured
Design Arena: spreadsheets986Not measuredOnly Qwen3.7 Max measured
LiveBench researchNot measured84.7%Only Qwen4 Max measured
LiveBench browserNot measured80.7%Only Qwen4 Max measured
LiveBench long agenticNot measured46.6%Only Qwen4 Max measured
Finance Agent (Vals AI)47.8%Not measuredOnly Qwen3.7 Max measured
Tax Agent Bench (Vals AI)15.8%40.4%Qwen4 Max
Legal Research Bench (Vals AI)25.5%41.4%Qwen4 Max
Harvey's Legal Agent Benchmark (Vals AI)Not measured21.3%Only Qwen4 Max measured
Excel Modeling Benchmark (Vals AI)57.0%Not measuredOnly Qwen3.7 Max measured
τ²-Bench Telecom94.7%Not measuredOnly Qwen3.7 Max measured
Neither model has a result on the other 26 boards here
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index11 boards, none shared
Reasoning benchmarks for Qwen3.7 Max and Qwen4 Max
BenchmarkQwen3.7 MaxQwen4 MaxLeader
GPQA Diamond90.2%Not measuredOnly Qwen3.7 Max measured
LiveBench reasoningNot measured92.2%Only Qwen4 Max measured
GPQA Diamond (Artificial Analysis run)92.3%Not measuredOnly Qwen3.7 Max measured
Humanity's Last Exam (Artificial Analysis run)40.5%Not measuredOnly Qwen3.7 Max measured
HLE-Diamond (Artificial Analysis run)Not measured44.9%Only Qwen4 Max measured
CritPt v2, hard subset (Artificial Analysis run)Not measured43.0%Only Qwen4 Max measured
GPQA Diamond (Epoch AI run)90.9%Not measuredOnly Qwen3.7 Max measured
SimpleBench70.4%Not measuredOnly Qwen3.7 Max measured
Chess puzzles27.0%Not measuredOnly Qwen3.7 Max measured
Mystery game puzzles32.0%Not measuredOnly Qwen3.7 Max measured
EBR-bench9.5%Not measuredOnly Qwen3.7 Max measured
Neither model has a result on the other 15 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
Math benchmarks for Qwen3.7 Max and Qwen4 Max
BenchmarkQwen3.7 MaxQwen4 MaxLeader
OTIS Mock AIME95.6%Not measuredOnly Qwen3.7 Max measured
FrontierMath, tiers 1 to 364.6%Not measuredOnly Qwen3.7 Max measured
FrontierMath, tier 434.1%Not measuredOnly Qwen3.7 Max measured
ProofBench26.0%Not measuredOnly Qwen3.7 Max measured
Neither model has a result on the other 10 boards here
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Knowledge benchmarks for Qwen3.7 Max and Qwen4 Max
BenchmarkQwen3.7 MaxQwen4 MaxLeader
MMLU Pro89.3%Not measuredOnly Qwen3.7 Max measured
LegalBench (Vals AI)84.9%Not measuredOnly Qwen3.7 Max measured
CorpFin (Vals AI)63.7%Not measuredOnly Qwen3.7 Max measured
TaxEval (Vals AI)75.3%Not measuredOnly Qwen3.7 Max measured
MedCode (Vals AI)38.8%Not measuredOnly Qwen3.7 Max measured
AA-OmniscienceNot measured27.7Only Qwen4 Max measured
SimpleQA Verified55.8%Not measuredOnly Qwen3.7 Max measured
Neither model has a result on the other 14 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
Web design benchmarks for Qwen3.7 Max and Qwen4 Max
BenchmarkQwen3.7 MaxQwen4 MaxLeader
Design Arena: websites1263Not measuredOnly Qwen3.7 Max measured
Design Arena: UI components1257Not measuredOnly Qwen3.7 Max measured
Design Arena: game dev1236Not measuredOnly Qwen3.7 Max measured
Design Arena: data viz1271Not measuredOnly Qwen3.7 Max measured
Design Arena: 3D1231Not measuredOnly Qwen3.7 Max measured
Design Arena: all code categories1257Not measuredOnly Qwen3.7 Max measured
Design Arena: SVG1200Not measuredOnly Qwen3.7 Max measured
Design Arena: ASCII art1195Not measuredOnly Qwen3.7 Max measured
LiveBench designNot measured79.1%Only Qwen4 Max measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index01
General benchmarks for Qwen3.7 Max and Qwen4 Max
BenchmarkQwen3.7 MaxQwen4 MaxLeader
Vals IndexNot measured63.2%Only Qwen4 Max measured
LiveBench data analysisNot measured83.4%Only Qwen4 Max measured
MedScribe (Vals AI)79.4%Not measuredOnly Qwen3.7 Max measured
IFBench80.5%Not measuredOnly Qwen3.7 Max measured
Long context reasoning (AA-LCR)79.0%Not measuredOnly Qwen3.7 Max measured
GDPval-AA11341809Qwen4 Max
DTBench92.3%Not measuredOnly Qwen3.7 Max measured
LMCA44.0%Not measuredOnly Qwen3.7 Max measured
EQ-Bench 41100Not measuredOnly Qwen3.7 Max 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 Qwen3.7 Max and Qwen4 Max
BenchmarkQwen3.7 MaxQwen4 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

Qwen3.7 Max$0.0037Fits in one request
Qwen4 Max$0.0036Fits in one request

Qwen4 Max costs less

Repository review

50K input + 3K output tokens

Qwen3.7 Max$0.087Fits in one request
Qwen4 Max$0.084Fits in one request

Qwen4 Max costs less

Agent loop

200K input + 10K output tokens

Qwen3.7 Max$0.339Fits in one request
Qwen4 Max$0.328Fits in one request

Qwen4 Max 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 Qwen3.7 Max and Qwen4 Max
SpecificationQwen3.7 MaxQwen4 Max
MakerQwenQwen
WeightsNo public weights listedNo public weights listed
ReleasedMay 21, 2026November 16, 2026
Context window1M1M
Input $/Mtok$1.475$1.425
Output $/Mtok$4.425$4.275
Coding index54.174.0
Independent measurements behind it34
Cost per solved task$4.69Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Qwen3.7 Max better than Qwen4 Max for coding?

Pick Qwen4 Max for coding: it scores 74.0 on the Vibe Coding Index against 54.1 for Qwen3.7 Max. Qwen3.7 Max is neither cheaper nor larger in context, and leads no benchmark both models were measured on.

Which is cheaper, Qwen3.7 Max or Qwen4 Max?

Qwen3.7 Max costs $1.475 per million input tokens and $4.425 per million output tokens; Qwen4 Max costs $1.425 and $4.275. Cost per solved task is known for Qwen3.7 Max only.

Which has the bigger context window, Qwen3.7 Max or Qwen4 Max?

Neither: both take 1M tokens.

How many benchmarks is the Qwen3.7 Max vs Qwen4 Max comparison based on?

10 benchmarks have a result for both models as of November 22, 2026. Qwen3.7 Max's coding index rests on 3 independent benchmarks and Qwen4 Max's on 4. 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/qwen3-7-max-vs-qwen4-max" title="Qwen3.7 Max vs Qwen4 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/qwen3-7-max-vs-qwen4-max">LLMBenchmark.ai, Qwen3.7 Max vs Qwen4 Max</a></p>