Claude Opus 4.7 vs Qwen3.8 27B
Updated October 21, 2026. 72 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick Qwen3.8 27B for coding: it scores 73.6 on the Vibe Coding Index against 51.1 for Claude Opus 4.7.
Pick Claude Opus 4.7 when what matters is AA Coding Index or LiveCodeBench (Vals AI), where it leads.
- Shared results
- 72
- Claude Opus 4.7 only
- 29
- Qwen3.8 27B only
- 13
- Like-for-like categories
- 7 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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, four public coding benchmarks
Qwen3.8 27B
Qwen3.8 27B scores 73.6 against 51.1.
1 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
Claude Opus 4.7
Claude Opus 4.7 leads, 40.7 against 33.7.
Same board, both measured
Agentic work
AA Agentic Index
Qwen3.8 27B
Qwen3.8 27B leads, 45.8 against 38.6.
Same board, both measured
Reasoning
GPQA Diamond
Claude Opus 4.7
Claude Opus 4.7 leads, 90.2% against 88.9%.
Same board, both measured
Math
AIME
No call
Only Claude Opus 4.7 has a AIME result, 96.3%.
Directional only
Knowledge
MMLU Pro
Claude Opus 4.7
Claude Opus 4.7 leads, 89.9% against 84.3%.
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 27B
Claude Opus 4.7 $0.018, Qwen3.8 27B $0.0007 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
Qwen3.8 27B
Claude Opus 4.7 $2.95, Qwen3.8 27B $1.26.
Published attempt cost and pass rate
At a glance
- Coding index
- Claude Opus 4.7 51.1 from 1 benchmark, Qwen3.8 27B 73.6 from 2 benchmarks.
- Widest category
- Math, 18.0 points apart on average over 3 shared benchmarks.
- Widest single gap
- ProofBench: Claude Opus 4.7 54.0% against Qwen3.8 27B 16.0%, 38.0 points apart on a 0 to 100 scale.
- Output price
- Claude Opus 4.7 $25.00 against Qwen3.8 27B $1.35 per million output tokens, 18.5x 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 Opus 4.7 | Bars | Qwen3.8 27B | Ties |
|---|---|---|---|---|
| Coding | 8 | 4 | 1 tie | |
| Intelligence | 2 | 0 | ||
| Agentic work | 4 | 2 | ||
| Reasoning | 11 | 0 | ||
| Math | 3 | 0 | ||
| Knowledge | 9 | 0 | ||
| General | 26 | 2 | ||
| All shared boards | 63 | 8 | 1 tie |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index84
| Benchmark | Claude Opus 4.7 | Qwen3.8 27B | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 86.8% | Only Qwen3.8 27B measured |
| LMArena WebDev | 1558 | 1593 | Qwen3.8 27B |
| AA Coding Index | 73.6 | 68.1 | Claude Opus 4.7 |
| LiveCodeBench (Vals AI) | 85.1% | 84.0% | Claude Opus 4.7 |
| SWE-bench Verified (Vals AI) | 82.0% | 86.0% | Qwen3.8 27B |
| Vibe Code Bench (Vals AI) | 71.0% | 64.8% | Claude Opus 4.7 |
| Code migration (Vals AI) | 43.9% | 14.2% | Claude Opus 4.7 |
| ProgramBench (Vals AI) | 0.0% | 0.0% | Tie |
| IOI (Vals AI) | Not measured | 39.1% | Only Qwen3.8 27B measured |
| Terminal-Bench 2.1 (Vals AI) | 68.5% | 58.4% | Claude Opus 4.7 |
| SkillsBench (Vals AI) | Not measured | 38.1% | Only Qwen3.8 27B measured |
| CyberBench (Vals AI) | Not measured | 67.8% | Only Qwen3.8 27B measured |
| SciCode | Not measured | 46.6% | Only Qwen3.8 27B measured |
| Terminal-Bench Hard (Artificial Analysis run) | 54.5% | Not measured | Only Claude Opus 4.7 measured |
| Terminal-Bench | 80.2% | Not measured | Only Claude Opus 4.7 measured |
| FrontierCode | 38.5% | Not measured | Only Claude Opus 4.7 measured |
| GSO-Bench | 44.1% | Not measured | Only Claude Opus 4.7 measured |
| WeirdML | 76.4% | Not measured | Only Claude Opus 4.7 measured |
| MirrorCode | 31.1% | Not measured | Only Claude Opus 4.7 measured |
| LMArena Text: coding | 1519 | 1483 | Claude Opus 4.7 |
| LMArena Text: software and IT | 1516 | 1478 | Claude Opus 4.7 |
| LMArena WebDev: React | 1560 | 1602 | Qwen3.8 27B |
| LMArena WebDev: HTML | 1554 | 1570 | Qwen3.8 27B |
| LMArena WebDev: from an image | 1577 | 1577 | Claude Opus 4.7 |
| Neither model has a result on the other 14 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Claude Opus 4.7 | Qwen3.8 27B | Leader |
|---|---|---|---|
| AA Intelligence Index | 40.7 | 33.7 | Claude Opus 4.7 |
| Epoch Capabilities Index | 156.8 | 151.7 | Claude Opus 4.7 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index42
| Benchmark | Claude Opus 4.7 | Qwen3.8 27B | Leader |
|---|---|---|---|
| AA Agentic Index | 38.6 | 45.8 | Qwen3.8 27B |
| LiveBench research | Not measured | 59.4% | Only Qwen3.8 27B measured |
| LiveBench browser | Not measured | 32.0% | Only Qwen3.8 27B measured |
| LiveBench long agentic | Not measured | 27.9% | Only Qwen3.8 27B measured |
| Finance Agent (Vals AI) | 51.5% | 48.5% | Claude Opus 4.7 |
| Tax Agent Bench (Vals AI) | 23.7% | 30.9% | Qwen3.8 27B |
| Legal Research Bench (Vals AI) | 38.5% | 36.1% | Claude Opus 4.7 |
| Excel Modeling Benchmark (Vals AI) | 63.7% | 59.7% | Claude Opus 4.7 |
| Terminal-Bench-Science (Vals AI) | Not measured | 0.0% | Only Qwen3.8 27B measured |
| τ²-Bench Telecom | 88.6% | Not measured | Only Claude Opus 4.7 measured |
| APEX-Agents | 49.2% | 47.5% | Claude Opus 4.7 |
| OSWorld 2.0 | 18.2% | Not measured | Only Claude Opus 4.7 measured |
| PostTrainBench | 28.6% | Not measured | Only Claude Opus 4.7 measured |
| Vending-Bench 2 | $10,937 | Not measured | Only Claude Opus 4.7 measured |
| LMArena Search | 1230 | Not measured | Only Claude Opus 4.7 measured |
| Neither model has a result on the other 24 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index110
| Benchmark | Claude Opus 4.7 | Qwen3.8 27B | Leader |
|---|---|---|---|
| GPQA Diamond | 90.2% | 88.9% | Claude Opus 4.7 |
| LiveBench reasoning | Not measured | 80.0% | Only Qwen3.8 27B measured |
| MMMU Pro (Vals AI) | 85.5% | 83.9% | Claude Opus 4.7 |
| GPQA Diamond (Artificial Analysis run) | 91.4% | 90.5% | Claude Opus 4.7 |
| Humanity's Last Exam (Artificial Analysis run) | 42.3% | 33.9% | Claude Opus 4.7 |
| GPQA Diamond (Epoch AI run) | 90.2% | Not measured | Only Claude Opus 4.7 measured |
| ARC-AGI | 93.5% | 87.5% | Claude Opus 4.7 |
| ARC-AGI-2 | 75.8% | 42.4% | Claude Opus 4.7 |
| SimpleBench | 61.7% | Not measured | Only Claude Opus 4.7 measured |
| Chess puzzles | 41.0% | Not measured | Only Claude Opus 4.7 measured |
| Mystery game puzzles | 28.0% | Not measured | Only Claude Opus 4.7 measured |
| EBR-bench | 19.0% | Not measured | Only Claude Opus 4.7 measured |
| LMArena Text: hard prompts | 1506 | 1459 | Claude Opus 4.7 |
| LMArena Text: expert | 1521 | 1479 | Claude Opus 4.7 |
| LMArena Text: hard prompts in English | 1508 | 1461 | Claude Opus 4.7 |
| LMArena Vision: diagrams | 1336 | 1285 | Claude Opus 4.7 |
| LMArena Vision: homework | 1337 | 1295 | Claude Opus 4.7 |
| Neither model has a result on the other 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index30
| Benchmark | Claude Opus 4.7 | Qwen3.8 27B | Leader |
|---|---|---|---|
| AIME | 96.3% | Not measured | Only Claude Opus 4.7 measured |
| OTIS Mock AIME | 97.8% | Not measured | Only Claude Opus 4.7 measured |
| FrontierMath, tiers 1 to 3 | 70.2% | Not measured | Only Claude Opus 4.7 measured |
| FrontierMath, tier 4 | 31.7% | Not measured | Only Claude Opus 4.7 measured |
| ProofBench | 54.0% | 16.0% | Claude Opus 4.7 |
| LMArena Text: math | 1498 | 1455 | Claude Opus 4.7 |
| LMArena Text: mathematical work | 1509 | 1458 | Claude Opus 4.7 |
| Neither model has a result on the other 6 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index90
| Benchmark | Claude Opus 4.7 | Qwen3.8 27B | Leader |
|---|---|---|---|
| MMLU Pro | 89.9% | 84.3% | Claude Opus 4.7 |
| LegalBench (Vals AI) | 85.3% | 82.4% | Claude Opus 4.7 |
| CaseLaw (Vals AI) | 68.4% | Not measured | Only Claude Opus 4.7 measured |
| CorpFin (Vals AI) | 66.1% | Not measured | Only Claude Opus 4.7 measured |
| TaxEval (Vals AI) | 75.3% | 70.8% | Claude Opus 4.7 |
| MortgageTax (Vals AI) | 70.3% | 64.9% | Claude Opus 4.7 |
| MedCode (Vals AI) | 54.9% | 28.7% | Claude Opus 4.7 |
| SimpleQA Verified | 51.7% | Not measured | Only Claude Opus 4.7 measured |
| LMArena Text: science | 1508 | 1467 | Claude Opus 4.7 |
| LMArena Text: medicine and healthcare | 1495 | 1437 | Claude Opus 4.7 |
| LMArena Text: legal and government | 1504 | 1447 | Claude Opus 4.7 |
| LMArena Text: business and finance | 1490 | 1438 | Claude Opus 4.7 |
| Neither model has a result on the other 5 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Claude Opus 4.7 | Qwen3.8 27B | Leader |
|---|---|---|---|
| LiveBench design | Not measured | 75.9% | Only Qwen3.8 27B measured |
| Neither model has a result on the other 8 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index262
| Benchmark | Claude Opus 4.7 | Qwen3.8 27B | Leader |
|---|---|---|---|
| LiveBench data analysis | Not measured | 76.6% | Only Qwen3.8 27B measured |
| SAGE (Vals AI) | 56.1% | 52.4% | Claude Opus 4.7 |
| MedScribe (Vals AI) | 83.0% | 83.8% | Qwen3.8 27B |
| Vals Multimodal Index | 67.4% | Not measured | Only Claude Opus 4.7 measured |
| IFBench | 58.6% | Not measured | Only Claude Opus 4.7 measured |
| Long context reasoning (AA-LCR) | 78.7% | 82.0% | Qwen3.8 27B |
| DTBench | 94.7% | 88.0% | Claude Opus 4.7 |
| LMCA | 52.2% | 41.4% | Claude Opus 4.7 |
| Surface Evolver Bench | Not measured | 45.0% | Only Qwen3.8 27B measured |
| Furniture assembly | 33.3% | Not measured | Only Claude Opus 4.7 measured |
| ExploitBench | 26.5% | Not measured | Only Claude Opus 4.7 measured |
| LMArena Text | 1490 | 1440 | Claude Opus 4.7 |
| LMArena Text: instruction following | 1498 | 1436 | Claude Opus 4.7 |
| LMArena Text: creative writing | 1486 | 1383 | Claude Opus 4.7 |
| LMArena Vision | 1317 | 1269 | Claude Opus 4.7 |
| LMArena Text: English | 1496 | 1448 | Claude Opus 4.7 |
| LMArena Text: non-English | 1480 | 1428 | Claude Opus 4.7 |
| LMArena Text: Chinese | 1531 | 1501 | Claude Opus 4.7 |
| LMArena Text: German | 1495 | 1443 | Claude Opus 4.7 |
| LMArena Text: Japanese | 1474 | 1387 | Claude Opus 4.7 |
| LMArena Text: Korean | 1464 | 1384 | Claude Opus 4.7 |
| LMArena Text: French | 1503 | 1465 | Claude Opus 4.7 |
| LMArena Text: Spanish | 1494 | 1449 | Claude Opus 4.7 |
| LMArena Text: Russian | 1494 | 1415 | Claude Opus 4.7 |
| LMArena Text: Polish | 1489 | 1439 | Claude Opus 4.7 |
| LMArena Text: multi-turn | 1505 | 1440 | Claude Opus 4.7 |
| LMArena Text: longer queries | 1505 | 1448 | Claude Opus 4.7 |
| LMArena Text: writing and language | 1492 | 1405 | Claude Opus 4.7 |
| LMArena Text: entertainment and media | 1474 | 1393 | Claude Opus 4.7 |
| LMArena Vision: English | 1322 | 1267 | Claude Opus 4.7 |
| LMArena Vision: Chinese | 1371 | 1326 | Claude Opus 4.7 |
| LMArena Vision: reading text | 1327 | 1281 | Claude Opus 4.7 |
| LMArena Vision: humor | 1309 | 1235 | Claude Opus 4.7 |
| LMArena Vision: recognizing things | 1278 | Not measured | Only Claude Opus 4.7 measured |
| LMArena Vision: creative writing | 1326 | 1237 | Claude Opus 4.7 |
| LMArena Document | 1495 | Not measured | Only Claude Opus 4.7 measured |
| Neither model has a result on the other 10 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Opus 4.7 | Qwen3.8 27B | 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 27B costs less
Repository review
50K input + 3K output tokens
Qwen3.8 27B costs less
Agent loop
200K input + 10K output tokens
Qwen3.8 27B 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 Opus 4.7 | Qwen3.8 27B |
|---|---|---|
| Maker | Anthropic | Qwen |
| Weights | No public weights listed | Open weights |
| Released | April 16, 2026 | August 14, 2026 |
| Context window | 1M | 1M |
| Input $/Mtok | $5.00 | $0.04 |
| Output $/Mtok | $25.00 | $1.35 |
| Coding index | 51.1 | 73.6 |
| Independent measurements behind it | 1 | 2 |
| Cost per solved task | $2.95 | $1.26 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Opus 4.7 better than Qwen3.8 27B for coding?
Pick Qwen3.8 27B for coding: it scores 73.6 on the Vibe Coding Index against 51.1 for Claude Opus 4.7. Pick Claude Opus 4.7 when what matters is AA Coding Index or LiveCodeBench (Vals AI), where it leads.
Which is cheaper, Claude Opus 4.7 or Qwen3.8 27B?
Claude Opus 4.7 costs $5.00 per million input tokens and $25.00 per million output tokens; Qwen3.8 27B costs $0.04 and $1.35. Per solved SWE-bench Verified task, Claude Opus 4.7 costs $2.95 and Qwen3.8 27B $1.26.
Which has the bigger context window, Claude Opus 4.7 or Qwen3.8 27B?
Neither: both take 1M tokens.
How many benchmarks is the Claude Opus 4.7 vs Qwen3.8 27B comparison based on?
72 benchmarks have a result for both models as of October 21, 2026. Claude Opus 4.7's coding index rests on 1 independent benchmark and Qwen3.8 27B's on 2. Where only one model has a result, the page says so instead of naming a winner.