Claude Fable 5 vs Claude Opus 4.6
Claude Fable 5 Verdict
- Claude Fable 5 leads on quality: 70.4 vs 63.1.
- Claude Fable 5 is stronger in agents & tools, coding, composite, human preference, instruction following, knowledge, long context, multimodal, reasoning.
- Claude Opus 4.6 is stronger in maths.
- Claude Opus 4.6 is 2.0× cheaper ($10.00 vs $20.00 per 1M blended).
- Claude Fable 5 streams 1.7× faster (63 vs 38 tokens per second).
| Metric | Claude Fable 5 | Claude Opus 4.6 |
|---|---|---|
| BenchLeader Index | 70.4 | 63.1 |
| Agents & tools score | 74.4 | 68.8 |
| Coding score | 70.6 | 62.3 |
| Composite score | 91.6 | 69.2 |
| Human preference score | 71.3 | 64.6 |
| Instruction following score | 62.6 | 53.7 |
| Knowledge score | 71.3 | 70.5 |
| Long context score | 67.8 | 65.5 |
| Multimodal score | 69.6 | 63.6 |
| Reasoning score | 75.6 | 62.3 |
| Maths score | – | 65.6 |
| Blended price $/M | $20.00 | $10.00 |
| Output speed | 63 tok/s | 38 tok/s |
| Time to first answer | 92.2 s | 2.0 s |
| Context window | 1M | 1M |
| GPQA Diamond | – | 90.5% |
| OTIS Mock AIME | – | 94.4% |
| SWE-bench Verified (Epoch) | – | 78.7% |
| Humanity's Last Exam | – | 19.0% |
| Terminal-Bench | – | 79.8% |
| SimpleBench | – | 67.6% |
| Cybench | – | 93.0% |
| Remote Labor Index | 16.1% | 4.2% |
| WeirdML | – | 77.9% |
| APEX-Agents | 45.0% | 32.4% |
| FrontierCode | 53.5% | 26.6% |
| GSO-Bench | – | 33.3% |
| Epoch Capabilities Index | 163.4 | 155.3 |
| LMArena Text | 1507 | 1498 |
| LMArena Hard Prompts | 1533 | 1527 |
| LMArena Coding | 1551 | 1546 |
| LMArena WebDev | 1628 | 1537 |
| LMArena Vision | 1330 | 1311 |
| AA Intelligence Index | 49.7 | 31.9 |
| IFBench | 63.5% | 53.1% |
| AA-LCR | 82.3% | 78.0% |
| MMMU-Pro | – | 75.4% |
| AA-Omniscience | 43.3 | 13.7 |
| Terminal-Bench Hard | 62.9% | 48.5% |
| GPQA Diamond (AA) | 92.6% | 89.6% |
| Humanity's Last Exam (AA) | 55.5% | 39.9% |
| SciCode (AA) | 61.0% | – |
| τ²-Bench Telecom (AA) | 98.5% | 92.1% |
| LiveCodeBench | 89.8% | – |
| MMLU-Pro | 91.5% | – |
| IOI | 72.3% | – |
| LegalBench | 88.6% | – |
| CorpFin | 71.8% | – |
| TaxEval | 76.9% | – |
| Terminal-Bench 2.1 (Vals) | 80.5% | – |
| SWE-bench (Vals) | 95.0% | – |
| GPQA Diamond (Vals) | 93.2% | – |
| Vals Index | 66.0 | – |
| MCP Atlas | 83.3% | – |
| PRBench Finance | 53.9% | – |
| PRBench Legal | 52.6% | – |
| HiL-Bench | 56.3% | 38.3% |
| EQ-Bench 4 | 1340 | 1223 |
| Kagi LLM Benchmark | 88.8% | 72.4% |
| SWE-bench Verified (bash only) | – | 75.6% |
| SWE-bench Verified (any scaffold) | – | 75.6% |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.