Enter opens the highlighted result. Esc closes.

Claude Opus 5 vs Qwen3.8 Max (0902)

Claude Opus 5 leads Qwen3.8 Max (0902) on coding, 83.0 to 62.4 on the Vibe Coding Index, and Qwen3.8 Max (0902) costs 4.2x less per output token ($6.00 vs $25.00 per million). Updated October 22, 2026; 112 benchmarks have a result for both.

Replace Claude Opus 5 Replace Qwen3.8 Max (0902)
Claude Opus 5Anthropic
Coding
83.0
Overall
81.6
Output $/M
$25.00
102
Qwen3.8 Max (0902)Qwen
Coding
62.4
Overall
68.9
Output $/M
$6.00
10
vs boards won of 112 shared

Claude Opus 5 or Qwen3.8 Max (0902): which to pick

Pick Claude Opus 5 for coding: it scores 83.0 on the Vibe Coding Index against 62.4 for Qwen3.8 Max (0902).

Pick Qwen3.8 Max (0902) for output tokens 4.2x cheaper ($6.00 vs $25.00 per million) and a lower cost per solved task ($1.32 against $1.33).

Model AClaude Opus 5Anthropic83.0/1006 independent benchmarks$1.33 per solved taskModel BQwen3.8 Max (0902)Qwen62.4/1005 independent benchmarks$1.32 per solved task
Shared results
112
Claude Opus 5 only
26
Qwen3.8 Max (0902) only
2
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
Claude Opus 5 83.0 from 6 benchmarks, Qwen3.8 Max (0902) 62.4 from 5 benchmarks.
Widest category
Math, 16.1 points apart on average over 7 shared benchmarks.
Widest single gap
LiveBench browser: Claude Opus 5 84.0% against Qwen3.8 Max (0902) 32.0%, 52.0 points apart on a 0 to 100 scale.
Output price
Claude Opus 5 $25.00 against Qwen3.8 Max (0902) $6.00 per million output tokens, 4.2x 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 5BarsQwen3.8 Max (0902)Ties
Coding210
Intelligence20
Agentic work152
Reasoning122
Math61
Knowledge101
Web design71
General293
All shared boards10210
CodingLike-for-likeSix of these boards feed the Vibe Coding Index210
Coding benchmarks for Claude Opus 5 and Qwen3.8 Max (0902)
BenchmarkClaude Opus 5Qwen3.8 Max (0902)Leader
LiveBench coding90.8%60.3%Claude Opus 5
LMArena WebDev16891674Claude Opus 5
AA Coding Index78.076.2Claude Opus 5
LiveCodeBench (Vals AI)89.0%87.8%Claude Opus 5
SWE-bench Verified (Vals AI)97.0%85.6%Claude Opus 5
Terminal-Bench 4 (Vals AI)53.5%34.3%Claude Opus 5
Vibe Code Bench (Vals AI)88.4%64.7%Claude Opus 5
Code migration (Vals AI)57.5%24.0%Claude Opus 5
ProgramBench (Vals AI)3.0%0.0%Claude Opus 5
IOI (Vals AI)84.3%68.9%Claude Opus 5
Terminal-Bench 2.1 (Vals AI)84.6%67.4%Claude Opus 5
SWE-bench Multimodal (Vals AI)Not measured42.5%Only Qwen3.8 Max (0902) measured
Vibe Code Bench 1-100 (Vals AI)28.5%12.8%Claude Opus 5
SkillsBench (Vals AI)60.4%42.0%Claude Opus 5
CyberBench (Vals AI)65.4%28.6%Claude Opus 5
SciCode56.4%53.2%Claude Opus 5
FrontierCode53.4%Not measuredOnly Claude Opus 5 measured
FrontierSWE52.0%17.8%Claude Opus 5
WeirdML91.8%Not measuredOnly Claude Opus 5 measured
ALE-Bench2164.6Not measuredOnly Claude Opus 5 measured
LMArena Text: coding15331502Claude Opus 5
LMArena Text: software and IT15271498Claude Opus 5
LMArena WebDev: React17111692Claude Opus 5
LMArena WebDev: HTML16801649Claude Opus 5
LMArena WebDev: from an image16621637Claude Opus 5
SciCode13.8%Not measuredOnly Claude Opus 5 measured
SWE-Bench Pro V299.4%Not measuredOnly Claude Opus 5 measured
SWE Atlas: Codebase QnA63.2%Not measuredOnly Claude Opus 5 measured
SWE Atlas: Test Writing62.2%Not measuredOnly Claude Opus 5 measured
Neither model has a result on the other 14 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Claude Opus 5 and Qwen3.8 Max (0902)
BenchmarkClaude Opus 5Qwen3.8 Max (0902)Leader
AA Intelligence Index50.845.4Claude Opus 5
Epoch Capabilities Index163.9157.4Claude Opus 5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index152
Agentic work benchmarks for Claude Opus 5 and Qwen3.8 Max (0902)
BenchmarkClaude Opus 5Qwen3.8 Max (0902)Leader
AA Agentic Index56.556.0Claude Opus 5
Design Arena: agents building web apps12661300Qwen3.8 Max (0902)
Design Arena: agents building full-stack apps13061305Claude Opus 5
Design Arena: agents building mobile apps13481224Claude Opus 5
Design Arena: agents building Android apps12561190Claude Opus 5
Design Arena: agents building Godot games12791234Claude Opus 5
Design Arena: agents building games1260Not measuredOnly Claude Opus 5 measured
Design Arena: PowerPoint slides12401232Claude Opus 5
Design Arena: spreadsheetsNot measured993Only Qwen3.8 Max (0902) measured
LiveBench research88.9%69.9%Claude Opus 5
LiveBench browser84.0%32.0%Claude Opus 5
LiveBench long agentic62.1%42.3%Claude Opus 5
SREBench (Vals AI)12.2%Not measuredOnly Claude Opus 5 measured
Finance Agent (Vals AI)58.6%50.6%Claude Opus 5
Tax Agent Bench (Vals AI)45.5%32.1%Claude Opus 5
Legal Research Bench (Vals AI)55.3%47.6%Claude Opus 5
Excel Modeling Benchmark (Vals AI)73.6%60.1%Claude Opus 5
BioMysteryBench (Vals AI)79.3%Not measuredOnly Claude Opus 5 measured
Terminal-Bench-Science (Vals AI)27.1%12.9%Claude Opus 5
Vals Science Index47.9%Not measuredOnly Claude Opus 5 measured
Time Horizon Index (Vals AI)18.8%Not measuredOnly Claude Opus 5 measured
CUA-bench (Vals AI)9.0%Not measuredOnly Claude Opus 5 measured
APEX-Agents65.8%63.3%Claude Opus 5
OSWorld 2.031.4%Not measuredOnly Claude Opus 5 measured
BALROG63.4%Not measuredOnly Claude Opus 5 measured
PostTrainBench35.0%Not measuredOnly Claude Opus 5 measured
Vending-Bench 2$11,182Not measuredOnly Claude Opus 5 measured
LMArena Search12421261Qwen3.8 Max (0902)
MCP Atlas85.8%Not measuredOnly Claude Opus 5 measured
Neither model has a result on the other 13 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index122
Reasoning benchmarks for Claude Opus 5 and Qwen3.8 Max (0902)
BenchmarkClaude Opus 5Qwen3.8 Max (0902)Leader
GPQA Diamond93.4%93.7%Qwen3.8 Max (0902)
LiveBench reasoning91.2%88.2%Claude Opus 5
MysteryMechanism (Vals AI)37.4%23.9%Claude Opus 5
MMMU Pro (Vals AI)89.9%88.0%Claude Opus 5
GPQA Diamond (Artificial Analysis run)93.7%92.8%Claude Opus 5
Humanity's Last Exam (Artificial Analysis run)54.9%43.1%Claude Opus 5
GPQA Diamond (Epoch AI run)93.9%92.7%Claude Opus 5
ARC-AGI97.5%Not measuredOnly Claude Opus 5 measured
ARC-AGI-290.4%Not measuredOnly Claude Opus 5 measured
SimpleBench80.6%Not measuredOnly Claude Opus 5 measured
Chess puzzles43.0%46.0%Qwen3.8 Max (0902)
Mystery game puzzles59.0%38.0%Claude Opus 5
EBR-bench45.7%Not measuredOnly Claude Opus 5 measured
LMArena Text: hard prompts15271496Claude Opus 5
LMArena Text: expert15541507Claude Opus 5
LMArena Text: hard prompts in English15231503Claude Opus 5
LMArena Vision: diagrams13481323Claude Opus 5
LMArena Vision: homework13381336Claude Opus 5
Humanity's Last Exam (Scale run)38.6%Not measuredOnly Claude Opus 5 measured
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index61
Math benchmarks for Claude Opus 5 and Qwen3.8 Max (0902)
BenchmarkClaude Opus 5Qwen3.8 Max (0902)Leader
OTIS Mock AIME98.9%100.0%Qwen3.8 Max (0902)
FrontierMath, tiers 1 to 385.6%74.7%Claude Opus 5
FrontierMath, tier 473.2%46.3%Claude Opus 5
ProofBench99.0%58.0%Claude Opus 5
LMArena Text: math15331501Claude Opus 5
LMArena Text: mathematical work15441491Claude Opus 5
MathArena70.4%51.6%Claude Opus 5
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index101
Knowledge benchmarks for Claude Opus 5 and Qwen3.8 Max (0902)
BenchmarkClaude Opus 5Qwen3.8 Max (0902)Leader
MMLU Pro91.6%88.6%Claude Opus 5
LegalBench (Vals AI)87.0%83.6%Claude Opus 5
CorpFin (Vals AI)73.2%65.8%Claude Opus 5
TaxEval (Vals AI)75.1%75.5%Qwen3.8 Max (0902)
MortgageTax (Vals AI)72.1%64.0%Claude Opus 5
MedCode (Vals AI)63.6%40.7%Claude Opus 5
SimpleQA Verified59.9%47.3%Claude Opus 5
LMArena Text: science15311512Claude Opus 5
LMArena Text: medicine and healthcare15251503Claude Opus 5
LMArena Text: legal and government15301492Claude Opus 5
LMArena Text: business and finance15011480Claude Opus 5
Neither model has a result on the other 9 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index71
Web design benchmarks for Claude Opus 5 and Qwen3.8 Max (0902)
BenchmarkClaude Opus 5Qwen3.8 Max (0902)Leader
Design Arena: websites13091275Claude Opus 5
Design Arena: UI components13461336Claude Opus 5
Design Arena: game dev13381312Claude Opus 5
Design Arena: data viz13301282Claude Opus 5
Design Arena: 3D13341336Qwen3.8 Max (0902)
Design Arena: all code categories13231292Claude Opus 5
Design Arena: SVG1309Not measuredOnly Claude Opus 5 measured
Design Arena: ASCII art13521200Claude Opus 5
LiveBench design83.0%78.6%Claude Opus 5
GeneralLike-for-likePublic boards, shown as published, kept out of the index293
General benchmarks for Claude Opus 5 and Qwen3.8 Max (0902)
BenchmarkClaude Opus 5Qwen3.8 Max (0902)Leader
LiveBench data analysis74.6%78.4%Qwen3.8 Max (0902)
Public Benefits Bench (Vals AI)76.9%67.1%Claude Opus 5
SAGE (Vals AI)49.4%51.2%Qwen3.8 Max (0902)
MedScribe (Vals AI)91.0%85.0%Claude Opus 5
Vals Multimodal Index73.9%65.4%Claude Opus 5
Long context reasoning (AA-LCR)82.0%80.3%Claude Opus 5
DTBench97.9%92.0%Claude Opus 5
LMCA64.5%46.2%Claude Opus 5
Furniture assembly60.8%20.0%Claude Opus 5
LMArena Text15091483Claude Opus 5
LMArena Text: instruction following15181481Claude Opus 5
LMArena Text: creative writing14931477Claude Opus 5
LMArena Vision13201313Claude Opus 5
LMArena Text: English15081491Claude Opus 5
LMArena Text: non-English15021473Claude Opus 5
LMArena Text: Chinese15731539Claude Opus 5
LMArena Text: German15271478Claude Opus 5
LMArena Text: Japanese15201473Claude Opus 5
LMArena Text: Korean15171460Claude Opus 5
LMArena Text: French15191498Claude Opus 5
LMArena Text: Spanish15231489Claude Opus 5
LMArena Text: Russian15071482Claude Opus 5
LMArena Text: Polish15221471Claude Opus 5
LMArena Text: multi-turn15001489Claude Opus 5
LMArena Text: longer queries15171488Claude Opus 5
LMArena Text: writing and language14991477Claude Opus 5
LMArena Text: entertainment and media14811464Claude Opus 5
LMArena Vision: English13151311Claude Opus 5
LMArena Vision: Chinese14001358Claude Opus 5
LMArena Vision: reading text13291322Claude Opus 5
LMArena Vision: humor12931304Qwen3.8 Max (0902)
LMArena Vision: creative writing13131312Claude Opus 5
LMArena Document1516Not measuredOnly Claude Opus 5 measured
EQ-Bench 41385Not measuredOnly Claude Opus 5 measured
Neither model has a result on the other 14 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Opus 5 and Qwen3.8 Max (0902)
BenchmarkClaude Opus 5Qwen3.8 Max (0902)Leader
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$0.018Fits in one request
Qwen3.8 Max (0902)$0.0050Fits in one request

Qwen3.8 Max (0902) costs less

Repository review

50K input + 3K output tokens

Claude Opus 5$0.325Fits in one request
Qwen3.8 Max (0902)$0.118Fits in one request

Qwen3.8 Max (0902) costs less

Agent loop

200K input + 10K output tokens

Claude Opus 5$1.25Fits in one request
Qwen3.8 Max (0902)$0.460Fits in one request

Qwen3.8 Max (0902) 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 and Qwen3.8 Max (0902)
SpecificationAnthropic Claude Opus 5Qwen3.8 Max (0902)
MakerAnthropicQwen
WeightsNo public weights listedNo public weights listed
ReleasedJuly 24, 2026September 3, 2026
Context window1M1M
Input $/Mtok$5.00$2.00
Output $/Mtok$25.00$6.00
Coding index83.062.4
Independent measurements behind it65
Cost per solved task$1.33$1.32
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Opus 5 better than Qwen3.8 Max (0902) for coding?

Pick Claude Opus 5 for coding: it scores 83.0 on the Vibe Coding Index against 62.4 for Qwen3.8 Max (0902). Pick Qwen3.8 Max (0902) for output tokens 4.2x cheaper ($6.00 vs $25.00 per million) and a lower cost per solved task ($1.32 against $1.33).

Which is cheaper, Claude Opus 5 or Qwen3.8 Max (0902)?

Claude Opus 5 costs $5.00 per million input tokens and $25.00 per million output tokens; Qwen3.8 Max (0902) costs $2.00 and $6.00. Per solved SWE-bench Verified task, Claude Opus 5 costs $1.33 and Qwen3.8 Max (0902) $1.32.

Which has the bigger context window, Claude Opus 5 or Qwen3.8 Max (0902)?

Neither: both take 1M tokens.

How many benchmarks is the Claude Opus 5 vs Qwen3.8 Max (0902) comparison based on?

112 benchmarks have a result for both models as of October 22, 2026. Claude Opus 5's coding index rests on 6 independent benchmarks and Qwen3.8 Max (0902)'s on 5. 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-vs-qwen3-8-max-0902" title="Claude Opus 5 vs Qwen3.8 Max (0902), 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-vs-qwen3-8-max-0902">LLMBenchmark.ai, Claude Opus 5 vs Qwen3.8 Max (0902)</a></p>