Claude Sonnet 5 vs Qwen3.8 Max (0902)
Qwen3.8 Max (0902) leads Claude Sonnet 5 on coding, 62.4 to 57.7 on the Vibe Coding Index, and Qwen3.8 Max (0902) costs 1.7x less per output token ($6.00 vs $10.00 per million). Updated October 22, 2026; 108 benchmarks have a result for both.
Claude Sonnet 5 or Qwen3.8 Max (0902): which to pick
Pick Qwen3.8 Max (0902) for coding: it scores 62.4 on the Vibe Coding Index against 57.7 for Claude Sonnet 5, though on 5 benchmarks against 6.
Pick Claude Sonnet 5 for work measured by LiveBench coding or Vibe Code Bench (Vals AI), where it leads.
- Shared results
- 108
- Claude Sonnet 5 only
- 13
- Qwen3.8 Max (0902) only
- 5
- 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.
Coding
Vibe Coding Index, six public coding benchmarks
Qwen3.8 Max (0902)
Qwen3.8 Max (0902) scores 62.4 against 57.7.
6 and 5 benchmarks behind the scores
Intelligence
AA Intelligence Index
Qwen3.8 Max (0902)
Qwen3.8 Max (0902) leads, 45.4 against 38.2.
Same board, both measured
Agentic work
AA Agentic Index
Qwen3.8 Max (0902)
Qwen3.8 Max (0902) leads, 56.0 against 43.6.
Same board, both measured
Reasoning
GPQA Diamond
Qwen3.8 Max (0902)
Qwen3.8 Max (0902) leads, 93.7% against 88.9%.
Same board, both measured
Knowledge
MMLU Pro
Qwen3.8 Max (0902)
Qwen3.8 Max (0902) leads, 88.6% against 87.5%.
Same board, both measured
Web design
Design Arena: websites
Qwen3.8 Max (0902)
Qwen3.8 Max (0902) leads, 1275 against 1274.
Same board, both measured
Long documents
Listed context window
No call
Both take 1M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3.8 Max (0902)
Claude Sonnet 5 $0.0070, Qwen3.8 Max (0902) $0.0050 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
Qwen3.8 Max (0902)
Claude Sonnet 5 $1.87, Qwen3.8 Max (0902) $1.32.
Published attempt cost and pass rate
At a glance
- Coding index
- Claude Sonnet 5 57.7 from 6 benchmarks, Qwen3.8 Max (0902) 62.4 from 5 benchmarks.
- Widest category
- Coding, 11.8 points apart on average over 20 shared benchmarks.
- Widest single gap
- CyberBench (Vals AI): Claude Sonnet 5 70.3% against Qwen3.8 Max (0902) 28.6%, 41.7 points apart on a 0 to 100 scale.
- Output price
- Claude Sonnet 5 $10.00 against Qwen3.8 Max (0902) $6.00 per million output tokens, 1.7x 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.
| Category | Claude Sonnet 5 | Bars | Qwen3.8 Max (0902) | Ties |
|---|---|---|---|---|
| Coding | 8 | 11 | 1 tie | |
| Intelligence | 0 | 2 | ||
| Agentic work | 6 | 11 | ||
| Reasoning | 1 | 12 | ||
| Math | 1 | 5 | ||
| Knowledge | 5 | 6 | ||
| Web design | 2 | 6 | ||
| General | 4 | 27 | ||
| All shared boards | 27 | 80 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index811
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| LiveBench coding | 74.7% | 60.3% | Claude Sonnet 5 |
| LMArena WebDev | 1540 | 1674 | Qwen3.8 Max (0902) |
| AA Coding Index | 71.5 | 76.2 | Qwen3.8 Max (0902) |
| LiveCodeBench (Vals AI) | 82.4% | 87.8% | Qwen3.8 Max (0902) |
| SWE-bench Verified (Vals AI) | 79.6% | 85.6% | Qwen3.8 Max (0902) |
| Terminal-Bench 4 (Vals AI) | 10.6% | 34.3% | Qwen3.8 Max (0902) |
| Vibe Code Bench (Vals AI) | 81.3% | 64.7% | Claude Sonnet 5 |
| Code migration (Vals AI) | 44.4% | 24.0% | Claude Sonnet 5 |
| ProgramBench (Vals AI) | 0.0% | 0.0% | Tie |
| IOI (Vals AI) | 45.0% | 68.9% | Qwen3.8 Max (0902) |
| Terminal-Bench 2.1 (Vals AI) | 74.5% | 67.4% | Claude Sonnet 5 |
| SWE-bench Multimodal (Vals AI) | Not measured | 42.5% | Only Qwen3.8 Max (0902) measured |
| Vibe Code Bench 1-100 (Vals AI) | 13.8% | 12.8% | Claude Sonnet 5 |
| SkillsBench (Vals AI) | 46.5% | 42.0% | Claude Sonnet 5 |
| CyberBench (Vals AI) | 70.3% | 28.6% | Claude Sonnet 5 |
| SciCode | 54.3% | 53.2% | Claude Sonnet 5 |
| FrontierCode | 42.7% | Not measured | Only Claude Sonnet 5 measured |
| FrontierSWE | Not measured | 17.8% | Only Qwen3.8 Max (0902) measured |
| GSO-Bench | 37.3% | Not measured | Only Claude Sonnet 5 measured |
| WeirdML | 68.8% | Not measured | Only Claude Sonnet 5 measured |
| ALE-Bench | 1463.1 | Not measured | Only Claude Sonnet 5 measured |
| LMArena Text: coding | 1483 | 1502 | Qwen3.8 Max (0902) |
| LMArena Text: software and IT | 1474 | 1498 | Qwen3.8 Max (0902) |
| LMArena WebDev: React | 1541 | 1692 | Qwen3.8 Max (0902) |
| LMArena WebDev: HTML | 1527 | 1649 | Qwen3.8 Max (0902) |
| LMArena WebDev: from an image | 1540 | 1637 | Qwen3.8 Max (0902) |
| SciCode | 10.7% | Not measured | Only Claude Sonnet 5 measured |
| Neither model has a result on the other 12 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| AA Intelligence Index | 38.2 | 45.4 | Qwen3.8 Max (0902) |
| Epoch Capabilities Index | 156.9 | 157.4 | Qwen3.8 Max (0902) |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index611
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| AA Agentic Index | 43.6 | 56.0 | Qwen3.8 Max (0902) |
| Design Arena: agents building web apps | 1245 | 1300 | Qwen3.8 Max (0902) |
| Design Arena: agents building full-stack apps | 1249 | 1305 | Qwen3.8 Max (0902) |
| Design Arena: agents building mobile apps | 1215 | 1224 | Qwen3.8 Max (0902) |
| Design Arena: agents building Android apps | 1227 | 1196 | Claude Sonnet 5 |
| Design Arena: agents building Godot games | 1225 | 1234 | Qwen3.8 Max (0902) |
| Design Arena: agents building games | 1227 | Not measured | Only Claude Sonnet 5 measured |
| Design Arena: HTML slides | 1168 | Not measured | Only Claude Sonnet 5 measured |
| Design Arena: PowerPoint slides | 1191 | 1233 | Qwen3.8 Max (0902) |
| Design Arena: spreadsheets | 1035 | 967 | Claude Sonnet 5 |
| LiveBench research | 82.8% | 69.9% | Claude Sonnet 5 |
| LiveBench browser | 72.0% | 32.0% | Claude Sonnet 5 |
| LiveBench long agentic | 39.0% | 42.3% | Qwen3.8 Max (0902) |
| SREBench (Vals AI) | 0.4% | Not measured | Only Claude Sonnet 5 measured |
| Finance Agent (Vals AI) | 53.9% | 50.6% | Claude Sonnet 5 |
| Tax Agent Bench (Vals AI) | 29.4% | 32.1% | Qwen3.8 Max (0902) |
| Legal Research Bench (Vals AI) | 41.8% | 47.6% | Qwen3.8 Max (0902) |
| Excel Modeling Benchmark (Vals AI) | 66.3% | 60.1% | Claude Sonnet 5 |
| Terminal-Bench-Science (Vals AI) | 5.7% | 12.9% | Qwen3.8 Max (0902) |
| APEX-Agents | 54.5% | 63.3% | Qwen3.8 Max (0902) |
| Vending-Bench 2 | $6,378 | Not measured | Only Claude Sonnet 5 measured |
| LMArena Search | Not measured | 1261 | Only Qwen3.8 Max (0902) measured |
| Neither model has a result on the other 19 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index112
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| GPQA Diamond | 88.9% | 93.7% | Qwen3.8 Max (0902) |
| LiveBench reasoning | 88.7% | 88.2% | Claude Sonnet 5 |
| MysteryMechanism (Vals AI) | Not measured | 23.9% | Only Qwen3.8 Max (0902) measured |
| MMMU Pro (Vals AI) | 83.0% | 88.0% | Qwen3.8 Max (0902) |
| GPQA Diamond (Artificial Analysis run) | 91.1% | 92.8% | Qwen3.8 Max (0902) |
| Humanity's Last Exam (Artificial Analysis run) | 41.3% | 43.1% | Qwen3.8 Max (0902) |
| GPQA Diamond (Epoch AI run) | 90.5% | 92.7% | Qwen3.8 Max (0902) |
| SimpleBench | 60.6% | Not measured | Only Claude Sonnet 5 measured |
| Chess puzzles | 45.0% | 46.0% | Qwen3.8 Max (0902) |
| Mystery game puzzles | 35.0% | 38.0% | Qwen3.8 Max (0902) |
| LMArena Text: hard prompts | 1461 | 1496 | Qwen3.8 Max (0902) |
| LMArena Text: expert | 1489 | 1507 | Qwen3.8 Max (0902) |
| LMArena Text: hard prompts in English | 1464 | 1503 | Qwen3.8 Max (0902) |
| LMArena Vision: diagrams | 1293 | 1323 | Qwen3.8 Max (0902) |
| LMArena Vision: homework | 1311 | 1336 | Qwen3.8 Max (0902) |
| Neither model has a result on the other 7 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index15
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| OTIS Mock AIME | 94.7% | 100.0% | Qwen3.8 Max (0902) |
| FrontierMath, tiers 1 to 3 | 65.6% | 74.7% | Qwen3.8 Max (0902) |
| FrontierMath, tier 4 | 29.3% | 46.3% | Qwen3.8 Max (0902) |
| ProofBench | 77.0% | 58.0% | Claude Sonnet 5 |
| LMArena Text: math | 1467 | 1501 | Qwen3.8 Max (0902) |
| LMArena Text: mathematical work | 1481 | 1491 | Qwen3.8 Max (0902) |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index56
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| MMLU Pro | 87.5% | 88.6% | Qwen3.8 Max (0902) |
| LegalBench (Vals AI) | 83.9% | 83.6% | Claude Sonnet 5 |
| CorpFin (Vals AI) | 67.0% | 65.8% | Claude Sonnet 5 |
| TaxEval (Vals AI) | 75.6% | 75.5% | Claude Sonnet 5 |
| MortgageTax (Vals AI) | 70.0% | 64.0% | Claude Sonnet 5 |
| MedCode (Vals AI) | 47.5% | 40.7% | Claude Sonnet 5 |
| SimpleQA Verified | 33.7% | 47.3% | Qwen3.8 Max (0902) |
| LMArena Text: science | 1460 | 1512 | Qwen3.8 Max (0902) |
| LMArena Text: medicine and healthcare | 1442 | 1503 | Qwen3.8 Max (0902) |
| LMArena Text: legal and government | 1454 | 1492 | Qwen3.8 Max (0902) |
| LMArena Text: business and finance | 1436 | 1480 | Qwen3.8 Max (0902) |
| Neither model has a result on the other 7 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index26
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| Design Arena: websites | 1274 | 1275 | Qwen3.8 Max (0902) |
| Design Arena: UI components | 1277 | 1338 | Qwen3.8 Max (0902) |
| Design Arena: game dev | 1277 | 1312 | Qwen3.8 Max (0902) |
| Design Arena: data viz | 1251 | 1281 | Qwen3.8 Max (0902) |
| Design Arena: 3D | 1261 | 1336 | Qwen3.8 Max (0902) |
| Design Arena: all code categories | 1275 | 1293 | Qwen3.8 Max (0902) |
| Design Arena: SVG | 1192 | Not measured | Only Claude Sonnet 5 measured |
| Design Arena: ASCII art | 1207 | 1200 | Claude Sonnet 5 |
| LiveBench design | 82.0% | 78.6% | Claude Sonnet 5 |
GeneralLike-for-likePublic boards, shown as published, kept out of the index427
| Benchmark | Claude Sonnet 5 | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| LiveBench data analysis | 71.7% | 78.4% | Qwen3.8 Max (0902) |
| Public Benefits Bench (Vals AI) | 66.0% | 67.1% | Qwen3.8 Max (0902) |
| SAGE (Vals AI) | 48.9% | 51.2% | Qwen3.8 Max (0902) |
| MedScribe (Vals AI) | 76.0% | 85.0% | Qwen3.8 Max (0902) |
| Vals Multimodal Index | 68.8% | 65.4% | Claude Sonnet 5 |
| Long context reasoning (AA-LCR) | 82.0% | 80.3% | Claude Sonnet 5 |
| DTBench | 92.5% | 92.0% | Claude Sonnet 5 |
| LMCA | 50.0% | 46.2% | Claude Sonnet 5 |
| Surface Evolver Bench | 60.0% | Not measured | Only Claude Sonnet 5 measured |
| Furniture assembly | Not measured | 20.0% | Only Qwen3.8 Max (0902) measured |
| LMArena Text | 1442 | 1483 | Qwen3.8 Max (0902) |
| LMArena Text: instruction following | 1452 | 1481 | Qwen3.8 Max (0902) |
| LMArena Text: creative writing | 1416 | 1477 | Qwen3.8 Max (0902) |
| LMArena Vision | 1273 | 1313 | Qwen3.8 Max (0902) |
| LMArena Text: English | 1450 | 1491 | Qwen3.8 Max (0902) |
| LMArena Text: non-English | 1431 | 1473 | Qwen3.8 Max (0902) |
| LMArena Text: Chinese | 1478 | 1539 | Qwen3.8 Max (0902) |
| LMArena Text: German | 1442 | 1478 | Qwen3.8 Max (0902) |
| LMArena Text: Japanese | 1422 | 1473 | Qwen3.8 Max (0902) |
| LMArena Text: Korean | 1412 | 1460 | Qwen3.8 Max (0902) |
| LMArena Text: French | 1463 | 1498 | Qwen3.8 Max (0902) |
| LMArena Text: Spanish | 1438 | 1489 | Qwen3.8 Max (0902) |
| LMArena Text: Russian | 1450 | 1482 | Qwen3.8 Max (0902) |
| LMArena Text: Polish | 1425 | 1471 | Qwen3.8 Max (0902) |
| LMArena Text: multi-turn | 1453 | 1489 | Qwen3.8 Max (0902) |
| LMArena Text: longer queries | 1462 | 1488 | Qwen3.8 Max (0902) |
| LMArena Text: writing and language | 1436 | 1477 | Qwen3.8 Max (0902) |
| LMArena Text: entertainment and media | 1411 | 1464 | Qwen3.8 Max (0902) |
| LMArena Vision: English | 1275 | 1311 | Qwen3.8 Max (0902) |
| LMArena Vision: Chinese | 1331 | 1358 | Qwen3.8 Max (0902) |
| LMArena Vision: reading text | 1288 | 1322 | Qwen3.8 Max (0902) |
| LMArena Vision: humor | 1261 | 1304 | Qwen3.8 Max (0902) |
| LMArena Vision: creative writing | 1246 | 1312 | Qwen3.8 Max (0902) |
| LMArena Document | 1466 | Not measured | Only Claude Sonnet 5 measured |
| Neither model has a result on the other 12 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Sonnet 5 | Qwen3.8 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
Qwen3.8 Max (0902) costs less
Repository review
50K input + 3K output tokens
Qwen3.8 Max (0902) costs less
Agent loop
200K input + 10K output tokens
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.
| Specification | Anthropic Claude Sonnet 5 | Qwen3.8 Max (0902) |
|---|---|---|
| Maker | Anthropic | Qwen |
| Weights | No public weights listed | No public weights listed |
| Released | June 30, 2026 | September 3, 2026 |
| Context window | 1M | 1M |
| Input $/Mtok | $2.00 | $2.00 |
| Output $/Mtok | $10.00 | $6.00 |
| Coding index | 57.7 | 62.4 |
| Independent measurements behind it | 6 | 5 |
| Cost per solved task | $1.87 | $1.32 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Sonnet 5 better than Qwen3.8 Max (0902) for coding?
Pick Qwen3.8 Max (0902) for coding: it scores 62.4 on the Vibe Coding Index against 57.7 for Claude Sonnet 5, though on 5 benchmarks against 6. Pick Claude Sonnet 5 for work measured by LiveBench coding or Vibe Code Bench (Vals AI), where it leads.
Which is cheaper, Claude Sonnet 5 or Qwen3.8 Max (0902)?
Claude Sonnet 5 costs $2.00 per million input tokens and $10.00 per million output tokens; Qwen3.8 Max (0902) costs $2.00 and $6.00. Per solved SWE-bench Verified task, Claude Sonnet 5 costs $1.87 and Qwen3.8 Max (0902) $1.32.
Which has the bigger context window, Claude Sonnet 5 or Qwen3.8 Max (0902)?
Neither: both take 1M tokens.
How many benchmarks is the Claude Sonnet 5 vs Qwen3.8 Max (0902) comparison based on?
108 benchmarks have a result for both models as of October 22, 2026. Claude Sonnet 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.