Claude Fable 5 vs Claude Opus 4.7
Claude Fable 5 leads Claude Opus 4.7 on coding, 79.3 to 67.5 on the Vibe Coding Index, and Claude Opus 4.7 costs 2.0x less per output token ($25.00 vs $50.00 per million). Updated October 22, 2026; 96 benchmarks have a result for both.
Claude Fable 5 or Claude Opus 4.7: which to pick
Pick Claude Fable 5 for coding: it scores 79.3 on the Vibe Coding Index against 67.5 for Claude Opus 4.7.
Pick Claude Opus 4.7 for output tokens 2.0x cheaper ($25.00 vs $50.00 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.
- Shared results
- 96
- Claude Fable 5 only
- 31
- Claude Opus 4.7 only
- 7
- 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.
Coding
Vibe Coding Index, six public coding benchmarks
Claude Fable 5
Claude Fable 5 scores 79.3 against 67.5.
6 and 4 benchmarks behind the scores
Intelligence
AA Intelligence Index
Claude Fable 5
Claude Fable 5 leads, 49.6 against 40.7.
Same board, both measured
Agentic work
AA Agentic Index
Claude Fable 5
Claude Fable 5 leads, 50.7 against 38.6.
Same board, both measured
Reasoning
GPQA Diamond
Claude Fable 5
Claude Fable 5 leads, 93.2% against 90.2%.
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 Fable 5
Claude Fable 5 leads, 91.5% against 89.9%.
Same board, both measured
Web design
Design Arena: websites
No call
Only Claude Fable 5 has a Design Arena: websites result, 1295.
Directional only
Long documents
Listed context window
No call
Both take 1M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Claude Opus 4.7
Claude Fable 5 $0.035, Claude Opus 4.7 $0.018 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
Claude Fable 5
Claude Fable 5 $2.16, Claude Opus 4.7 $2.95.
Published attempt cost and pass rate
At a glance
- Coding index
- Claude Fable 5 79.3 from 6 benchmarks, Claude Opus 4.7 67.5 from 4 benchmarks.
- Widest category
- Math, 20.4 points apart on average over 6 shared benchmarks.
- Widest single gap
- FrontierMath, tier 4: Claude Fable 5 90.2% against Claude Opus 4.7 31.7%, 58.5 points apart on a 0 to 100 scale.
- Output price
- Claude Fable 5 $50.00 against Claude Opus 4.7 $25.00 per million output tokens, 2.0x 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 Fable 5 | Bars | Claude Opus 4.7 | Ties |
|---|---|---|---|---|
| Coding | 16 | 2 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 8 | 2 | ||
| Reasoning | 16 | 1 | ||
| Math | 6 | 0 | ||
| Knowledge | 6 | 5 | ||
| General | 29 | 3 | ||
| All shared boards | 83 | 13 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index162
| Benchmark | Claude Fable 5 | Claude Opus 4.7 | Leader |
|---|---|---|---|
| LiveBench coding | 98.3% | Not measured | Only Claude Fable 5 measured |
| LMArena WebDev | 1624 | 1558 | Claude Fable 5 |
| AA Coding Index | 76.5 | 73.6 | Claude Fable 5 |
| LiveCodeBench (Vals AI) | 89.8% | 85.1% | Claude Fable 5 |
| SWE-bench Verified (Vals AI) | 95.0% | 82.0% | Claude Fable 5 |
| Terminal-Bench 4 (Vals AI) | 41.4% | Not measured | Only Claude Fable 5 measured |
| Vibe Code Bench (Vals AI) | 90.3% | 71.0% | Claude Fable 5 |
| Code migration (Vals AI) | 55.1% | 43.9% | Claude Fable 5 |
| ProgramBench (Vals AI) | 2.0% | 0.0% | Claude Fable 5 |
| Terminal-Bench 2.1 (Vals AI) | 80.5% | 68.5% | Claude Fable 5 |
| SciCode | 61.0% | Not measured | Only Claude Fable 5 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 62.9% | 54.5% | Claude Fable 5 |
| Terminal-Bench | Not measured | 80.2% | Only Claude Opus 4.7 measured |
| FrontierCode | 53.5% | 38.5% | Claude Fable 5 |
| FrontierSWE | 47.0% | Not measured | Only Claude Fable 5 measured |
| GSO-Bench | 78.4% | 44.1% | Claude Fable 5 |
| WeirdML | 91.9% | 76.4% | Claude Fable 5 |
| MirrorCode | 63.9% | 31.1% | Claude Fable 5 |
| ALE-Bench | 2041.3 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: coding | 1518 | 1519 | Claude Opus 4.7 |
| LMArena Text: software and IT | 1513 | 1516 | Claude Opus 4.7 |
| LMArena WebDev: React | 1631 | 1560 | Claude Fable 5 |
| LMArena WebDev: HTML | 1651 | 1554 | Claude Fable 5 |
| LMArena WebDev: from an image | 1623 | 1577 | Claude Fable 5 |
| Neither model has a result on the other 15 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Claude Fable 5 | Claude Opus 4.7 | Leader |
|---|---|---|---|
| AA Intelligence Index | 49.6 | 40.7 | Claude Fable 5 |
| Epoch Capabilities Index | 163.4 | 156.8 | Claude Fable 5 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index82
| Benchmark | Claude Fable 5 | Claude Opus 4.7 | Leader |
|---|---|---|---|
| AA Agentic Index | 50.7 | 38.6 | Claude Fable 5 |
| Design Arena: agents building web apps | 1261 | Not measured | Only Claude Fable 5 measured |
| Design Arena: agents building full-stack apps | 1280 | Not measured | Only Claude Fable 5 measured |
| Design Arena: agents building mobile apps | 1264 | Not measured | Only Claude Fable 5 measured |
| Design Arena: agents building Android apps | 1284 | Not measured | Only Claude Fable 5 measured |
| Design Arena: agents building Godot games | 1292 | Not measured | Only Claude Fable 5 measured |
| Design Arena: agents building games | 1285 | Not measured | Only Claude Fable 5 measured |
| Design Arena: HTML slides | 1227 | Not measured | Only Claude Fable 5 measured |
| Design Arena: PowerPoint slides | 1270 | Not measured | Only Claude Fable 5 measured |
| LiveBench research | 86.9% | Not measured | Only Claude Fable 5 measured |
| LiveBench browser | 92.0% | Not measured | Only Claude Fable 5 measured |
| LiveBench long agentic | 53.2% | Not measured | Only Claude Fable 5 measured |
| Finance Agent (Vals AI) | 56.3% | 51.5% | Claude Fable 5 |
| Tax Agent Bench (Vals AI) | 38.3% | 23.7% | Claude Fable 5 |
| Legal Research Bench (Vals AI) | 49.5% | 38.5% | Claude Fable 5 |
| Excel Modeling Benchmark (Vals AI) | 73.7% | 63.7% | Claude Fable 5 |
| Terminal-Bench-Science (Vals AI) | 15.7% | Not measured | Only Claude Fable 5 measured |
| τ²-Bench Telecom | 98.5% | 88.6% | Claude Fable 5 |
| APEX-Agents | 63.6% | 49.2% | Claude Fable 5 |
| Remote Labor Index | 16.1% | Not measured | Only Claude Fable 5 measured |
| OSWorld 2.0 | Not measured | 18.2% | Only Claude Opus 4.7 measured |
| PostTrainBench | 41.8% | 28.6% | Claude Fable 5 |
| Vending-Bench 2 | $5,680 | $10,937 | Claude Opus 4.7 |
| LMArena Search | 1229 | 1230 | Claude Opus 4.7 |
| Neither model has a result on the other 17 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index161
| Benchmark | Claude Fable 5 | Claude Opus 4.7 | Leader |
|---|---|---|---|
| GPQA Diamond | 93.2% | 90.2% | Claude Fable 5 |
| LiveBench reasoning | 89.7% | Not measured | Only Claude Fable 5 measured |
| MMMU Pro (Vals AI) | 89.3% | 85.5% | Claude Fable 5 |
| GPQA Diamond (Artificial Analysis run) | 92.6% | 91.4% | Claude Fable 5 |
| Humanity's Last Exam (Artificial Analysis run) | 55.5% | 42.3% | Claude Fable 5 |
| GPQA Diamond (Epoch AI run) | 85.9% | 90.2% | Claude Opus 4.7 |
| ARC-AGI | 98.5% | 93.5% | Claude Fable 5 |
| ARC-AGI-2 | 89.2% | 75.8% | Claude Fable 5 |
| SimpleBench | 81.9% | 61.7% | Claude Fable 5 |
| Chess puzzles | 47.0% | 41.0% | Claude Fable 5 |
| Mystery game puzzles | 52.0% | 28.0% | Claude Fable 5 |
| EBR-bench | 39.5% | 19.0% | Claude Fable 5 |
| LMArena Text: hard prompts | 1507 | 1506 | Claude Fable 5 |
| LMArena Text: expert | 1534 | 1521 | Claude Fable 5 |
| LMArena Text: hard prompts in English | 1509 | 1508 | Claude Fable 5 |
| LMArena Vision: diagrams | 1357 | 1336 | Claude Fable 5 |
| LMArena Vision: homework | 1361 | 1337 | Claude Fable 5 |
| Kagi LLM Benchmark | 91.4% | 80.7% | Claude Fable 5 |
| Neither model has a result on the other 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | Claude Fable 5 | Claude Opus 4.7 | Leader |
|---|---|---|---|
| AIME | Not measured | 96.3% | Only Claude Opus 4.7 measured |
| OTIS Mock AIME | 100.0% | 97.8% | Claude Fable 5 |
| FrontierMath, tiers 1 to 3 | 87.0% | 70.2% | Claude Fable 5 |
| FrontierMath, tier 4 | 90.2% | 31.7% | Claude Fable 5 |
| ProofBench | 95.0% | 54.0% | Claude Fable 5 |
| LMArena Text: math | 1516 | 1498 | Claude Fable 5 |
| LMArena Text: mathematical work | 1515 | 1509 | Claude Fable 5 |
| Neither model has a result on the other 6 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index65
| Benchmark | Claude Fable 5 | Claude Opus 4.7 | Leader |
|---|---|---|---|
| MMLU Pro | 91.5% | 89.9% | Claude Fable 5 |
| LegalBench (Vals AI) | 88.6% | 85.3% | Claude Fable 5 |
| CaseLaw (Vals AI) | Not measured | 68.4% | Only Claude Opus 4.7 measured |
| CorpFin (Vals AI) | 71.8% | 66.1% | Claude Fable 5 |
| TaxEval (Vals AI) | 76.9% | 75.3% | Claude Fable 5 |
| MortgageTax (Vals AI) | 68.9% | 70.3% | Claude Opus 4.7 |
| MedCode (Vals AI) | 56.1% | 54.9% | Claude Fable 5 |
| SimpleQA Verified | 70.7% | 51.7% | Claude Fable 5 |
| LMArena Text: science | 1506 | 1508 | Claude Opus 4.7 |
| LMArena Text: medicine and healthcare | 1480 | 1495 | Claude Opus 4.7 |
| LMArena Text: legal and government | 1498 | 1504 | Claude Opus 4.7 |
| LMArena Text: business and finance | 1480 | 1490 | Claude Opus 4.7 |
| FACTS Benchmark Suite | Not measured | 45.5% | Only Claude Opus 4.7 measured |
| Neither model has a result on the other 5 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | Claude Fable 5 | Claude Opus 4.7 | Leader |
|---|---|---|---|
| Design Arena: websites | 1295 | Not measured | Only Claude Fable 5 measured |
| Design Arena: UI components | 1316 | Not measured | Only Claude Fable 5 measured |
| Design Arena: game dev | 1335 | Not measured | Only Claude Fable 5 measured |
| Design Arena: data viz | 1304 | Not measured | Only Claude Fable 5 measured |
| Design Arena: 3D | 1318 | Not measured | Only Claude Fable 5 measured |
| Design Arena: all code categories | 1310 | Not measured | Only Claude Fable 5 measured |
| Design Arena: SVG | 1302 | Not measured | Only Claude Fable 5 measured |
| Design Arena: ASCII art | 1329 | Not measured | Only Claude Fable 5 measured |
| LiveBench design | 84.7% | Not measured | Only Claude Fable 5 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index293
| Benchmark | Claude Fable 5 | Claude Opus 4.7 | Leader |
|---|---|---|---|
| LiveBench data analysis | 80.5% | Not measured | Only Claude Fable 5 measured |
| Public Benefits Bench (Vals AI) | 70.4% | Not measured | Only Claude Fable 5 measured |
| SAGE (Vals AI) | 51.9% | 56.1% | Claude Opus 4.7 |
| MedScribe (Vals AI) | 88.5% | 83.0% | Claude Fable 5 |
| Vals Multimodal Index | 74.2% | 67.4% | Claude Fable 5 |
| IFBench | 63.5% | 58.6% | Claude Fable 5 |
| Long context reasoning (AA-LCR) | 82.3% | 78.7% | Claude Fable 5 |
| DTBench | 98.4% | 94.7% | Claude Fable 5 |
| LMCA | 61.2% | 52.2% | Claude Fable 5 |
| Surface Evolver Bench | 95.0% | Not measured | Only Claude Fable 5 measured |
| Furniture assembly | 35.8% | 33.3% | Claude Fable 5 |
| ExploitBench | Not measured | 26.5% | Only Claude Opus 4.7 measured |
| LMArena Text | 1491 | 1490 | Claude Fable 5 |
| LMArena Text: instruction following | 1502 | 1498 | Claude Fable 5 |
| LMArena Text: creative writing | 1494 | 1486 | Claude Fable 5 |
| LMArena Vision | 1325 | 1317 | Claude Fable 5 |
| LMArena Text: English | 1498 | 1496 | Claude Fable 5 |
| LMArena Text: non-English | 1482 | 1480 | Claude Fable 5 |
| LMArena Text: Chinese | 1543 | 1531 | Claude Fable 5 |
| LMArena Text: German | 1486 | 1495 | Claude Opus 4.7 |
| LMArena Text: Japanese | 1504 | 1474 | Claude Fable 5 |
| LMArena Text: Korean | 1487 | 1464 | Claude Fable 5 |
| LMArena Text: French | 1505 | 1503 | Claude Fable 5 |
| LMArena Text: Spanish | 1498 | 1494 | Claude Fable 5 |
| LMArena Text: Russian | 1503 | 1494 | Claude Fable 5 |
| LMArena Text: Polish | 1494 | 1489 | Claude Fable 5 |
| LMArena Text: multi-turn | 1504 | 1505 | Claude Opus 4.7 |
| LMArena Text: longer queries | 1510 | 1505 | Claude Fable 5 |
| LMArena Text: writing and language | 1501 | 1492 | Claude Fable 5 |
| LMArena Text: entertainment and media | 1482 | 1474 | Claude Fable 5 |
| LMArena Vision: English | 1323 | 1322 | Claude Fable 5 |
| LMArena Vision: Chinese | 1373 | 1371 | Claude Fable 5 |
| LMArena Vision: reading text | 1337 | 1327 | Claude Fable 5 |
| LMArena Vision: humor | 1341 | 1309 | Claude Fable 5 |
| LMArena Vision: recognizing things | Not measured | 1278 | Only Claude Opus 4.7 measured |
| LMArena Vision: creative writing | 1338 | 1326 | Claude Fable 5 |
| LMArena Document | 1496 | 1495 | Claude Fable 5 |
| Neither model has a result on the other 9 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Fable 5 | Claude Opus 4.7 | 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 4.7 costs less
Repository review
50K input + 3K output tokens
Claude Opus 4.7 costs less
Agent loop
200K input + 10K output tokens
Claude Opus 4.7 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 Fable 5 | Anthropic Claude Opus 4.7 |
|---|---|---|
| Maker | Anthropic | Anthropic |
| Weights | No public weights listed | No public weights listed |
| Released | June 9, 2026 | April 16, 2026 |
| Context window | 1M | 1M |
| Input $/Mtok | $10.00 | $5.00 |
| Output $/Mtok | $50.00 | $25.00 |
| Coding index | 79.3 | 67.5 |
| Independent measurements behind it | 6 | 4 |
| Cost per solved task | $2.16 | $2.95 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Fable 5 better than Claude Opus 4.7 for coding?
Pick Claude Fable 5 for coding: it scores 79.3 on the Vibe Coding Index against 67.5 for Claude Opus 4.7. Pick Claude Opus 4.7 for output tokens 2.0x cheaper ($25.00 vs $50.00 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.
Which is cheaper, Claude Fable 5 or Claude Opus 4.7?
Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens; Claude Opus 4.7 costs $5.00 and $25.00. Per solved SWE-bench Verified task, Claude Fable 5 costs $2.16 and Claude Opus 4.7 $2.95.
Which has the bigger context window, Claude Fable 5 or Claude Opus 4.7?
Neither: both take 1M tokens.
How many benchmarks is the Claude Fable 5 vs Claude Opus 4.7 comparison based on?
96 benchmarks have a result for both models as of October 22, 2026. Claude Fable 5's coding index rests on 6 independent benchmarks and Claude Opus 4.7's on 4. Where only one model has a result, the page says so instead of naming a winner.