Enter opens the highlighted result. Esc closes.

Claude Opus 5.5 vs Qwen4 Max

Claude Opus 5.5 leads Qwen4 Max on coding, 87.4 to 74.0 on the Vibe Coding Index, and Qwen4 Max costs 4.7x less per output token ($4.275 vs $20.00 per million). Updated November 22, 2026; 26 benchmarks have a result for both.

Claude Opus 5.5Anthropic
Coding
87.4
Overall
86.9
Output $/M
$20.00
23
Qwen4 MaxQwen
Coding
74.0
Overall
66.6
Output $/M
$4.275
2
vs boards won of 26 shared, 1 tied

Claude Opus 5.5 or Qwen4 Max: which to pick

Pick Claude Opus 5.5 for coding: it scores 87.4 on the Vibe Coding Index against 74.0 for Qwen4 Max.

Pick Qwen4 Max for output tokens 4.7x cheaper ($4.275 vs $20.00 per million) and work measured by Harvey's Legal Agent Benchmark (Vals AI) or LiveBench data analysis, where it leads.

Shared results
26
Claude Opus 5.5 only
112
Qwen4 Max only
0
Like-for-like categories
7 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
Claude Opus 5.5 87.4 from 5 benchmarks, Qwen4 Max 74.0 from 4 benchmarks.
Widest category
Intelligence, 16.2 points apart on average over 1 shared benchmark.
Widest single gap
LiveBench long agentic: Claude Opus 5.5 72.8% against Qwen4 Max 46.6%, 26.2 points apart on a 0 to 100 scale.
Output price
Claude Opus 5.5 $20.00 against Qwen4 Max $4.275 per million output tokens, 4.7x 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
CategoryClaude Opus 5.5BarsQwen4 MaxTies
Coding110
Intelligence10
Agentic work51
Reasoning201 tie
Knowledge10
Web design10
General21
All shared boards2321 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index110
Coding benchmarks for Claude Opus 5.5 and Qwen4 Max
BenchmarkClaude Opus 5.5Qwen4 MaxLeader
LiveBench coding87.1%69.8%Claude Opus 5.5
LMArena WebDev18191751Claude Opus 5.5
Terminal-Bench 4 (Vals AI)65.2%48.5%Claude Opus 5.5
Vibe Code Bench (Vals AI)90.3%86.7%Claude Opus 5.5
Code migration (Vals AI)66.7%Not measuredOnly Claude Opus 5.5 measured
ProgramBench (Vals AI)18.5%Not measuredOnly Claude Opus 5.5 measured
IOI (Vals AI)95.1%Not measuredOnly Claude Opus 5.5 measured
Terminal-Bench 2.1 (Vals AI)87.6%Not measuredOnly Claude Opus 5.5 measured
SWE-bench Multimodal (Vals AI)70.2%Not measuredOnly Claude Opus 5.5 measured
Vibe Code Bench 1-100 (Vals AI)30.4%Not measuredOnly Claude Opus 5.5 measured
CyberBench (Vals AI)74.6%Not measuredOnly Claude Opus 5.5 measured
SciCode66.9%Not measuredOnly Claude Opus 5.5 measured
Terminal-Bench 4.0 (Artificial Analysis run)59.6%48.5%Claude Opus 5.5
FrontierCode 2.046.4%26.5%Claude Opus 5.5
FrontierCode54.6%Not measuredOnly Claude Opus 5.5 measured
FrontierSWE62.3%Not measuredOnly Claude Opus 5.5 measured
MirrorCode95.6%Not measuredOnly Claude Opus 5.5 measured
ALE-Bench2147.41831.0Claude Opus 5.5
LMArena Text: coding1543Not measuredOnly Claude Opus 5.5 measured
LMArena Text: software and IT1533Not measuredOnly Claude Opus 5.5 measured
LMArena WebDev: React18711785Claude Opus 5.5
LMArena WebDev: HTML17621752Claude Opus 5.5
LMArena WebDev: from an image17581675Claude Opus 5.5
SciCode18.5%Not measuredOnly Claude Opus 5.5 measured
GameBench 295.082.4Claude Opus 5.5
Terminal-Bench 4.064.8%Not measuredOnly Claude Opus 5.5 measured
CursorBench 4.058.1%Not measuredOnly Claude Opus 5.5 measured
ProgramBench18.5%Not measuredOnly Claude Opus 5.5 measured
Neither model has a result on the other 24 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for Claude Opus 5.5 and Qwen4 Max
BenchmarkClaude Opus 5.5Qwen4 MaxLeader
AA Intelligence Index59.142.9Claude Opus 5.5
Epoch Capabilities Index168.7Not measuredOnly Claude Opus 5.5 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index51
Agentic work benchmarks for Claude Opus 5.5 and Qwen4 Max
BenchmarkClaude Opus 5.5Qwen4 MaxLeader
Design Arena: agents building web apps1308Not measuredOnly Claude Opus 5.5 measured
Design Arena: agents building full-stack apps1385Not measuredOnly Claude Opus 5.5 measured
Design Arena: agents building mobile apps1296Not measuredOnly Claude Opus 5.5 measured
Design Arena: agents building Android apps1402Not measuredOnly Claude Opus 5.5 measured
Design Arena: agents building Godot games1242Not measuredOnly Claude Opus 5.5 measured
Design Arena: agents building games1408Not measuredOnly Claude Opus 5.5 measured
Design Arena: HTML slides1475Not measuredOnly Claude Opus 5.5 measured
Design Arena: PowerPoint slides1378Not measuredOnly Claude Opus 5.5 measured
Design Arena: spreadsheets1289Not measuredOnly Claude Opus 5.5 measured
LiveBench research88.9%84.7%Claude Opus 5.5
LiveBench browser88.4%80.7%Claude Opus 5.5
LiveBench long agentic72.8%46.6%Claude Opus 5.5
SREBench (Vals AI)33.6%Not measuredOnly Claude Opus 5.5 measured
Finance Agent (Vals AI)58.6%Not measuredOnly Claude Opus 5.5 measured
Tax Agent Bench (Vals AI)44.5%40.4%Claude Opus 5.5
Legal Research Bench (Vals AI)50.5%41.4%Claude Opus 5.5
Harvey's Legal Agent Benchmark (Vals AI)9.2%21.3%Qwen4 Max
Excel Modeling Benchmark (Vals AI)75.9%Not measuredOnly Claude Opus 5.5 measured
BioMysteryBench (Vals AI)79.3%Not measuredOnly Claude Opus 5.5 measured
Terminal-Bench-Science (Vals AI)47.1%Not measuredOnly Claude Opus 5.5 measured
Vals Science Index58.7%Not measuredOnly Claude Opus 5.5 measured
Time Horizon Index (Vals AI)91.3%Not measuredOnly Claude Opus 5.5 measured
CUA-bench (Vals AI)14.0%Not measuredOnly Claude Opus 5.5 measured
Robo-bench (Vals AI)14.6%Not measuredOnly Claude Opus 5.5 measured
APEX-Agents73.5%Not measuredOnly Claude Opus 5.5 measured
Remote Labor Index21.3%Not measuredOnly Claude Opus 5.5 measured
PostTrainBench43.8%Not measuredOnly Claude Opus 5.5 measured
SpireClimb5.8%Not measuredOnly Claude Opus 5.5 measured
Vending-Bench 2$9,235Not measuredOnly Claude Opus 5.5 measured
LMArena Search1214Not measuredOnly Claude Opus 5.5 measured
MCP Atlas85.4%Not measuredOnly Claude Opus 5.5 measured
Neither model has a result on the other 13 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index20
Reasoning benchmarks for Claude Opus 5.5 and Qwen4 Max
BenchmarkClaude Opus 5.5Qwen4 MaxLeader
LiveBench reasoning92.2%92.2%Tie
MysteryMechanism (Vals AI)49.5%Not measuredOnly Claude Opus 5.5 measured
Humanity's Last Exam (Artificial Analysis run)61.4%Not measuredOnly Claude Opus 5.5 measured
HLE-Diamond (Artificial Analysis run)67.9%44.9%Claude Opus 5.5
CritPt v2, hard subset (Artificial Analysis run)68.0%43.0%Claude Opus 5.5
GPQA Diamond (Epoch AI run)90.6%Not measuredOnly Claude Opus 5.5 measured
ARC-AGI98.5%Not measuredOnly Claude Opus 5.5 measured
ARC-AGI-293.3%Not measuredOnly Claude Opus 5.5 measured
SimpleBench88.4%Not measuredOnly Claude Opus 5.5 measured
Chess puzzles67.0%Not measuredOnly Claude Opus 5.5 measured
Mystery game puzzles71.0%Not measuredOnly Claude Opus 5.5 measured
EBR-bench71.4%Not measuredOnly Claude Opus 5.5 measured
LMArena Text: hard prompts1531Not measuredOnly Claude Opus 5.5 measured
LMArena Text: expert1558Not measuredOnly Claude Opus 5.5 measured
LMArena Text: hard prompts in English1534Not measuredOnly Claude Opus 5.5 measured
LMArena Vision: diagrams1352Not measuredOnly Claude Opus 5.5 measured
LMArena Vision: homework1363Not measuredOnly Claude Opus 5.5 measured
Humanity's Last Exam (Scale run)55.0%Not measuredOnly Claude Opus 5.5 measured
GoBench2273Not measuredOnly Claude Opus 5.5 measured
Neither model has a result on the other 7 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Math benchmarks for Claude Opus 5.5 and Qwen4 Max
BenchmarkClaude Opus 5.5Qwen4 MaxLeader
OTIS Mock AIME100.0%Not measuredOnly Claude Opus 5.5 measured
FrontierMath, tiers 1 to 391.2%Not measuredOnly Claude Opus 5.5 measured
FrontierMath, tier 495.0%Not measuredOnly Claude Opus 5.5 measured
ProofBench100.0%Not measuredOnly Claude Opus 5.5 measured
LMArena Text: math1523Not measuredOnly Claude Opus 5.5 measured
LMArena Text: mathematical work1537Not measuredOnly Claude Opus 5.5 measured
MathArena75.3%Not measuredOnly Claude Opus 5.5 measured
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index10
Knowledge benchmarks for Claude Opus 5.5 and Qwen4 Max
BenchmarkClaude Opus 5.5Qwen4 MaxLeader
MedCode (Vals AI)49.8%Not measuredOnly Claude Opus 5.5 measured
AA-Omniscience48.127.7Claude Opus 5.5
SimpleQA Verified72.2%Not measuredOnly Claude Opus 5.5 measured
LMArena Text: science1524Not measuredOnly Claude Opus 5.5 measured
LMArena Text: medicine and healthcare1515Not measuredOnly Claude Opus 5.5 measured
LMArena Text: legal and government1523Not measuredOnly Claude Opus 5.5 measured
LMArena Text: business and finance1496Not measuredOnly Claude Opus 5.5 measured
FACTS Benchmark Suite66.5%Not measuredOnly Claude Opus 5.5 measured
Neither model has a result on the other 13 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index10
Web design benchmarks for Claude Opus 5.5 and Qwen4 Max
BenchmarkClaude Opus 5.5Qwen4 MaxLeader
Design Arena: websites1321Not measuredOnly Claude Opus 5.5 measured
Design Arena: UI components1368Not measuredOnly Claude Opus 5.5 measured
Design Arena: game dev1435Not measuredOnly Claude Opus 5.5 measured
Design Arena: data viz1383Not measuredOnly Claude Opus 5.5 measured
Design Arena: 3D1450Not measuredOnly Claude Opus 5.5 measured
Design Arena: all code categories1382Not measuredOnly Claude Opus 5.5 measured
Design Arena: SVG1315Not measuredOnly Claude Opus 5.5 measured
Design Arena: ASCII art1343Not measuredOnly Claude Opus 5.5 measured
LiveBench design83.3%79.1%Claude Opus 5.5
GeneralLike-for-likePublic boards, shown as published, kept out of the index21
General benchmarks for Claude Opus 5.5 and Qwen4 Max
BenchmarkClaude Opus 5.5Qwen4 MaxLeader
Vals Index70.1%63.2%Claude Opus 5.5
LiveBench data analysis80.3%83.4%Qwen4 Max
Public Benefits Bench (Vals AI)70.6%Not measuredOnly Claude Opus 5.5 measured
SAGE (Vals AI)91.3%Not measuredOnly Claude Opus 5.5 measured
MedScribe (Vals AI)91.4%Not measuredOnly Claude Opus 5.5 measured
Long context reasoning (AA-LCR)84.7%Not measuredOnly Claude Opus 5.5 measured
GDPval-AA18661809Claude Opus 5.5
DTBench98.9%Not measuredOnly Claude Opus 5.5 measured
LMCA68.2%Not measuredOnly Claude Opus 5.5 measured
Furniture assembly83.3%Not measuredOnly Claude Opus 5.5 measured
LMArena Text1513Not measuredOnly Claude Opus 5.5 measured
LMArena Text: instruction following1533Not measuredOnly Claude Opus 5.5 measured
LMArena Text: creative writing1531Not measuredOnly Claude Opus 5.5 measured
LMArena Vision1322Not measuredOnly Claude Opus 5.5 measured
LMArena Text: English1519Not measuredOnly Claude Opus 5.5 measured
LMArena Text: non-English1503Not measuredOnly Claude Opus 5.5 measured
LMArena Text: Chinese1575Not measuredOnly Claude Opus 5.5 measured
LMArena Text: German1525Not measuredOnly Claude Opus 5.5 measured
LMArena Text: Japanese1534Not measuredOnly Claude Opus 5.5 measured
LMArena Text: Korean1509Not measuredOnly Claude Opus 5.5 measured
LMArena Text: French1506Not measuredOnly Claude Opus 5.5 measured
LMArena Text: Spanish1514Not measuredOnly Claude Opus 5.5 measured
LMArena Text: Russian1514Not measuredOnly Claude Opus 5.5 measured
LMArena Text: Polish1515Not measuredOnly Claude Opus 5.5 measured
LMArena Text: multi-turn1502Not measuredOnly Claude Opus 5.5 measured
LMArena Text: longer queries1528Not measuredOnly Claude Opus 5.5 measured
LMArena Text: writing and language1528Not measuredOnly Claude Opus 5.5 measured
LMArena Text: entertainment and media1507Not measuredOnly Claude Opus 5.5 measured
LMArena Vision: English1326Not measuredOnly Claude Opus 5.5 measured
LMArena Vision: Chinese1376Not measuredOnly Claude Opus 5.5 measured
LMArena Vision: reading text1331Not measuredOnly Claude Opus 5.5 measured
LMArena Vision: humor1295Not measuredOnly Claude Opus 5.5 measured
LMArena Vision: creative writing1324Not measuredOnly Claude Opus 5.5 measured
EQ-Bench 41340Not measuredOnly Claude Opus 5.5 measured
Neither model has a result on the other 15 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Opus 5.5 and Qwen4 Max
BenchmarkClaude Opus 5.5Qwen4 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

Claude Opus 5.5$0.014Fits in one request
Qwen4 Max$0.0036Fits in one request

Qwen4 Max costs less

Repository review

50K input + 3K output tokens

Claude Opus 5.5$0.260Fits in one request
Qwen4 Max$0.084Fits in one request

Qwen4 Max costs less

Agent loop

200K input + 10K output tokens

Claude Opus 5.5$1.00Fits 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 Claude Opus 5.5 and Qwen4 Max
SpecificationAnthropic Claude Opus 5.5Qwen4 Max
MakerAnthropicQwen
WeightsNo public weights listedNo public weights listed
ReleasedSeptember 22, 2026November 16, 2026
Context window1M1M
Input $/Mtok$4.00$1.425
Output $/Mtok$20.00$4.275
Coding index87.474.0
Independent measurements behind it54
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Opus 5.5 better than Qwen4 Max for coding?

Pick Claude Opus 5.5 for coding: it scores 87.4 on the Vibe Coding Index against 74.0 for Qwen4 Max. Pick Qwen4 Max for output tokens 4.7x cheaper ($4.275 vs $20.00 per million) and work measured by Harvey's Legal Agent Benchmark (Vals AI) or LiveBench data analysis, where it leads.

Which is cheaper, Claude Opus 5.5 or Qwen4 Max?

Claude Opus 5.5 costs $4.00 per million input tokens and $20.00 per million output tokens; Qwen4 Max costs $1.425 and $4.275. Cost per solved task is known for neither of them.

Which has the bigger context window, Claude Opus 5.5 or Qwen4 Max?

Neither: both take 1M tokens.

How many benchmarks is the Claude Opus 5.5 vs Qwen4 Max comparison based on?

26 benchmarks have a result for both models as of November 22, 2026. Claude Opus 5.5's coding index rests on 5 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/claude-opus-5-5-vs-qwen4-max" title="Claude Opus 5.5 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/claude-opus-5-5-vs-qwen4-max">LLMBenchmark.ai, Claude Opus 5.5 vs Qwen4 Max</a></p>