Claude Opus 4.7 vs Claude Sonnet 4.6
Updated October 21, 2026. 90 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick Claude Sonnet 4.6 for coding: it scores 57.8 on the Vibe Coding Index against 51.1 for Claude Opus 4.7.
Pick Claude Opus 4.7 when what matters is LMArena WebDev or AA Coding Index, where it leads.
- Shared results
- 90
- Claude Opus 4.7 only
- 11
- Claude Sonnet 4.6 only
- 27
- 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
Claude Sonnet 4.6
Claude Sonnet 4.6 scores 57.8 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 30.1.
Same board, both measured
Agentic work
AA Agentic Index
Claude Opus 4.7
Claude Opus 4.7 leads, 38.6 against 31.8.
Same board, both measured
Reasoning
GPQA Diamond
Claude Opus 4.7
Claude Opus 4.7 leads, 90.2% against 85.6%.
Same board, both measured
Math
AIME
Claude Opus 4.7
Claude Opus 4.7 leads, 96.3% against 92.3%.
Same board, both measured
Knowledge
MMLU Pro
Claude Opus 4.7
Claude Opus 4.7 leads, 89.9% against 87.3%.
Same board, both measured
Web design
Design Arena: websites
No call
Only Claude Sonnet 4.6 has a Design Arena: websites result, 1285.
Directional only
Long documents
Listed context window
No call
Both take 1M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Claude Sonnet 4.6
Claude Opus 4.7 $0.018, Claude Sonnet 4.6 $0.011 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
Claude Sonnet 4.6
Claude Opus 4.7 $2.95, Claude Sonnet 4.6 $1.68.
Published attempt cost and pass rate
At a glance
- Coding index
- Claude Opus 4.7 51.1 from 1 benchmark, Claude Sonnet 4.6 57.8 from 2 benchmarks.
- Widest category
- Coding, 7.9 points apart on average over 17 shared benchmarks.
- Widest single gap
- Terminal-Bench: Claude Opus 4.7 80.2% against Claude Sonnet 4.6 53.4%, 26.8 points apart on a 0 to 100 scale.
- Output price
- Claude Opus 4.7 $25.00 against Claude Sonnet 4.6 $15.00 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 Opus 4.7 | Bars | Claude Sonnet 4.6 | Ties |
|---|---|---|---|---|
| Coding | 16 | 1 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 7 | 1 | 1 tie | |
| Reasoning | 14 | 0 | ||
| Math | 5 | 0 | ||
| Knowledge | 10 | 1 | ||
| General | 31 | 1 | ||
| All shared boards | 85 | 4 | 1 tie |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index161
| Benchmark | Claude Opus 4.7 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 66.6% | Only Claude Sonnet 4.6 measured |
| LMArena WebDev | 1558 | 1523 | Claude Opus 4.7 |
| AA Coding Index | 73.6 | 63.0 | Claude Opus 4.7 |
| LiveCodeBench (Vals AI) | 85.1% | 82.1% | Claude Opus 4.7 |
| SWE-bench Verified (Vals AI) | 82.0% | 77.4% | Claude Opus 4.7 |
| Vibe Code Bench (Vals AI) | 71.0% | 55.8% | Claude Opus 4.7 |
| Code migration (Vals AI) | 43.9% | 39.9% | Claude Opus 4.7 |
| ProgramBench (Vals AI) | 0.0% | 0.5% | Claude Sonnet 4.6 |
| Terminal-Bench 2.1 (Vals AI) | 68.5% | 57.3% | Claude Opus 4.7 |
| SkillsBench (Vals AI) | Not measured | 49.0% | Only Claude Sonnet 4.6 measured |
| SciCode | Not measured | 50.1% | Only Claude Sonnet 4.6 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 54.5% | 53.0% | Claude Opus 4.7 |
| Terminal-Bench | 80.2% | 53.4% | Claude Opus 4.7 |
| FrontierCode | 38.5% | 24.3% | Claude Opus 4.7 |
| GSO-Bench | 44.1% | Not measured | Only Claude Opus 4.7 measured |
| WeirdML | 76.4% | 66.1% | Claude Opus 4.7 |
| MirrorCode | 31.1% | Not measured | Only Claude Opus 4.7 measured |
| LMArena Text: coding | 1519 | 1503 | Claude Opus 4.7 |
| LMArena Text: software and IT | 1516 | 1491 | Claude Opus 4.7 |
| LMArena WebDev: React | 1560 | 1512 | Claude Opus 4.7 |
| LMArena WebDev: HTML | 1554 | 1527 | Claude Opus 4.7 |
| LMArena WebDev: from an image | 1577 | 1529 | Claude Opus 4.7 |
| Neither model has a result on the other 16 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Claude Opus 4.7 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| AA Intelligence Index | 40.7 | 30.1 | Claude Opus 4.7 |
| Epoch Capabilities Index | 156.8 | 152.8 | Claude Opus 4.7 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index71
| Benchmark | Claude Opus 4.7 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| AA Agentic Index | 38.6 | 31.8 | Claude Opus 4.7 |
| Design Arena: agents building web apps | Not measured | 1190 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building full-stack apps | Not measured | 1213 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building mobile apps | Not measured | 1212 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building Android apps | Not measured | 1202 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building Godot games | Not measured | 1226 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building games | Not measured | 1177 | Only Claude Sonnet 4.6 measured |
| LiveBench research | Not measured | 63.6% | Only Claude Sonnet 4.6 measured |
| LiveBench browser | Not measured | 68.0% | Only Claude Sonnet 4.6 measured |
| LiveBench long agentic | Not measured | 23.1% | Only Claude Sonnet 4.6 measured |
| Finance Agent (Vals AI) | 51.5% | 51.0% | Claude Opus 4.7 |
| Tax Agent Bench (Vals AI) | 23.7% | 25.8% | Claude Sonnet 4.6 |
| Legal Research Bench (Vals AI) | 38.5% | 38.5% | Tie |
| Excel Modeling Benchmark (Vals AI) | 63.7% | 60.2% | Claude Opus 4.7 |
| τ²-Bench Telecom | 88.6% | 79.5% | Claude Opus 4.7 |
| APEX-Agents | 49.2% | 43.0% | Claude Opus 4.7 |
| OSWorld 2.0 | 18.2% | 9.3% | Claude Opus 4.7 |
| DeepResearch Bench | Not measured | 54.9% | Only Claude Sonnet 4.6 measured |
| PostTrainBench | 28.6% | Not measured | Only Claude Opus 4.7 measured |
| OSWorld | Not measured | 72.1% | Only Claude Sonnet 4.6 measured |
| Vending-Bench 2 | $10,937 | $7,204 | Claude Opus 4.7 |
| LMArena Search | 1230 | Not measured | Only Claude Opus 4.7 measured |
| Neither model has a result on the other 17 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index140
| Benchmark | Claude Opus 4.7 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| GPQA Diamond | 90.2% | 85.6% | Claude Opus 4.7 |
| LiveBench reasoning | Not measured | 86.4% | Only Claude Sonnet 4.6 measured |
| MMMU Pro (Vals AI) | 85.5% | 83.6% | Claude Opus 4.7 |
| GPQA Diamond (Artificial Analysis run) | 91.4% | 87.5% | Claude Opus 4.7 |
| Humanity's Last Exam (Artificial Analysis run) | 42.3% | 33.6% | Claude Opus 4.7 |
| GPQA Diamond (Epoch AI run) | 90.2% | 87.4% | Claude Opus 4.7 |
| ARC-AGI | 93.5% | 86.5% | Claude Opus 4.7 |
| ARC-AGI-2 | 75.8% | 60.4% | Claude Opus 4.7 |
| SimpleBench | 61.7% | Not measured | Only Claude Opus 4.7 measured |
| Chess puzzles | 41.0% | 23.0% | Claude Opus 4.7 |
| Mystery game puzzles | 28.0% | 16.0% | Claude Opus 4.7 |
| EBR-bench | 19.0% | Not measured | Only Claude Opus 4.7 measured |
| LMArena Text: hard prompts | 1506 | 1485 | Claude Opus 4.7 |
| LMArena Text: expert | 1521 | 1500 | Claude Opus 4.7 |
| LMArena Text: hard prompts in English | 1508 | 1493 | Claude Opus 4.7 |
| LMArena Vision: diagrams | 1336 | 1306 | Claude Opus 4.7 |
| LMArena Vision: homework | 1337 | 1311 | 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 index50
| Benchmark | Claude Opus 4.7 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| AIME | 96.3% | 92.3% | Claude Opus 4.7 |
| OTIS Mock AIME | 97.8% | 85.8% | Claude Opus 4.7 |
| 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% | 45.0% | Claude Opus 4.7 |
| LMArena Text: math | 1498 | 1462 | Claude Opus 4.7 |
| LMArena Text: mathematical work | 1509 | 1483 | 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 index101
| Benchmark | Claude Opus 4.7 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| MMLU Pro | 89.9% | 87.3% | Claude Opus 4.7 |
| LegalBench (Vals AI) | 85.3% | 82.1% | Claude Opus 4.7 |
| CaseLaw (Vals AI) | 68.4% | 64.0% | Claude Opus 4.7 |
| CorpFin (Vals AI) | 66.1% | 65.3% | Claude Opus 4.7 |
| TaxEval (Vals AI) | 75.3% | 77.1% | Claude Sonnet 4.6 |
| MortgageTax (Vals AI) | 70.3% | 67.7% | Claude Opus 4.7 |
| MedQA (Vals AI) | Not measured | 92.1% | Only Claude Sonnet 4.6 measured |
| MedCode (Vals AI) | 54.9% | Not measured | Only Claude Opus 4.7 measured |
| SimpleQA Verified | 51.7% | 35.5% | Claude Opus 4.7 |
| LMArena Text: science | 1508 | 1474 | Claude Opus 4.7 |
| LMArena Text: medicine and healthcare | 1495 | 1451 | Claude Opus 4.7 |
| LMArena Text: legal and government | 1504 | 1470 | Claude Opus 4.7 |
| LMArena Text: business and finance | 1490 | 1464 | Claude Opus 4.7 |
| Neither model has a result on the other 4 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | Claude Opus 4.7 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1285 | Only Claude Sonnet 4.6 measured |
| Design Arena: UI components | Not measured | 1271 | Only Claude Sonnet 4.6 measured |
| Design Arena: game dev | Not measured | 1259 | Only Claude Sonnet 4.6 measured |
| Design Arena: data viz | Not measured | 1278 | Only Claude Sonnet 4.6 measured |
| Design Arena: 3D | Not measured | 1234 | Only Claude Sonnet 4.6 measured |
| Design Arena: all code categories | Not measured | 1276 | Only Claude Sonnet 4.6 measured |
| Design Arena: SVG | Not measured | 1196 | Only Claude Sonnet 4.6 measured |
| Design Arena: ASCII art | Not measured | 1235 | Only Claude Sonnet 4.6 measured |
| LiveBench design | Not measured | 71.8% | Only Claude Sonnet 4.6 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index311
| Benchmark | Claude Opus 4.7 | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| LiveBench data analysis | Not measured | 76.1% | Only Claude Sonnet 4.6 measured |
| Public Benefits Bench (Vals AI) | Not measured | 62.5% | Only Claude Sonnet 4.6 measured |
| SAGE (Vals AI) | 56.1% | 46.6% | Claude Opus 4.7 |
| MedScribe (Vals AI) | 83.0% | Not measured | Only Claude Opus 4.7 measured |
| Vals Multimodal Index | 67.4% | 60.6% | Claude Opus 4.7 |
| IFBench | 58.6% | 56.6% | Claude Opus 4.7 |
| Long context reasoning (AA-LCR) | 78.7% | 80.0% | Claude Sonnet 4.6 |
| DTBench | 94.7% | 89.9% | Claude Opus 4.7 |
| LMCA | 52.2% | 46.5% | Claude Opus 4.7 |
| Furniture assembly | 33.3% | Not measured | Only Claude Opus 4.7 measured |
| ExploitBench | 26.5% | 23.6% | Claude Opus 4.7 |
| LMArena Text | 1490 | 1458 | Claude Opus 4.7 |
| LMArena Text: instruction following | 1498 | 1476 | Claude Opus 4.7 |
| LMArena Text: creative writing | 1486 | 1435 | Claude Opus 4.7 |
| LMArena Vision | 1317 | 1283 | Claude Opus 4.7 |
| LMArena Text: English | 1496 | 1472 | Claude Opus 4.7 |
| LMArena Text: non-English | 1480 | 1440 | Claude Opus 4.7 |
| LMArena Text: Chinese | 1531 | 1491 | Claude Opus 4.7 |
| LMArena Text: German | 1495 | 1428 | Claude Opus 4.7 |
| LMArena Text: Japanese | 1474 | 1420 | Claude Opus 4.7 |
| LMArena Text: Korean | 1464 | 1410 | Claude Opus 4.7 |
| LMArena Text: French | 1503 | 1465 | Claude Opus 4.7 |
| LMArena Text: Spanish | 1494 | 1464 | Claude Opus 4.7 |
| LMArena Text: Russian | 1494 | 1440 | Claude Opus 4.7 |
| LMArena Text: Polish | 1489 | 1443 | Claude Opus 4.7 |
| LMArena Text: multi-turn | 1505 | 1464 | Claude Opus 4.7 |
| LMArena Text: longer queries | 1505 | 1479 | Claude Opus 4.7 |
| LMArena Text: writing and language | 1492 | 1444 | Claude Opus 4.7 |
| LMArena Text: entertainment and media | 1474 | 1431 | Claude Opus 4.7 |
| LMArena Vision: English | 1322 | 1298 | Claude Opus 4.7 |
| LMArena Vision: Chinese | 1371 | 1307 | Claude Opus 4.7 |
| LMArena Vision: reading text | 1327 | 1296 | Claude Opus 4.7 |
| LMArena Vision: humor | 1309 | 1248 | Claude Opus 4.7 |
| LMArena Vision: recognizing things | 1278 | 1195 | Claude Opus 4.7 |
| LMArena Vision: creative writing | 1326 | 1272 | Claude Opus 4.7 |
| LMArena Document | 1495 | 1482 | Claude Opus 4.7 |
| 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 | Claude Sonnet 4.6 | 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 Sonnet 4.6 costs less
Repository review
50K input + 3K output tokens
Claude Sonnet 4.6 costs less
Agent loop
200K input + 10K output tokens
Claude Sonnet 4.6 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 | Anthropic Claude Sonnet 4.6 |
|---|---|---|
| Maker | Anthropic | Anthropic |
| Weights | No public weights listed | No public weights listed |
| Released | April 16, 2026 | February 17, 2026 |
| Context window | 1M | 1M |
| Input $/Mtok | $5.00 | $3.00 |
| Output $/Mtok | $25.00 | $15.00 |
| Coding index | 51.1 | 57.8 |
| Independent measurements behind it | 1 | 2 |
| Cost per solved task | $2.95 | $1.68 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Opus 4.7 better than Claude Sonnet 4.6 for coding?
Pick Claude Sonnet 4.6 for coding: it scores 57.8 on the Vibe Coding Index against 51.1 for Claude Opus 4.7. Pick Claude Opus 4.7 when what matters is LMArena WebDev or AA Coding Index, where it leads.
Which is cheaper, Claude Opus 4.7 or Claude Sonnet 4.6?
Claude Opus 4.7 costs $5.00 per million input tokens and $25.00 per million output tokens; Claude Sonnet 4.6 costs $3.00 and $15.00. Per solved SWE-bench Verified task, Claude Opus 4.7 costs $2.95 and Claude Sonnet 4.6 $1.68.
Which has the bigger context window, Claude Opus 4.7 or Claude Sonnet 4.6?
Neither: both take 1M tokens.
How many benchmarks is the Claude Opus 4.7 vs Claude Sonnet 4.6 comparison based on?
90 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 Claude Sonnet 4.6's on 2. Where only one model has a result, the page says so instead of naming a winner.