Claude Fable 5.1 vs Claude Opus 4.6
Claude Fable 5.1 Verdict
- Claude Fable 5.1 (xhigh) leads on quality: 70.3 vs 63.1.
- Claude Fable 5.1 (xhigh) is stronger in coding, composite, knowledge, long context, reasoning.
- Claude Opus 4.6 is stronger in agents & tools, human preference, instruction following, maths, multimodal.
- Claude Opus 4.6 is 2.0× cheaper ($10.00 vs $20.00 per 1M blended).
- Claude Fable 5.1 (xhigh) streams 1.5× faster (58 vs 38 tokens per second).
| Metric | Claude Fable 5.1 (xhigh) | Claude Opus 4.6 |
|---|---|---|
| BenchLeader Index | 70.3 | 63.1 |
| Coding score | 71.3 | 62.3 |
| Composite score | 95.0 | 69.2 |
| Knowledge score | 84.1 | 70.5 |
| Long context score | 68.1 | 65.5 |
| Reasoning score | 76.4 | 62.3 |
| Agents & tools score | – | 68.8 |
| Human preference score | – | 64.6 |
| Instruction following score | – | 53.7 |
| Maths score | – | 65.6 |
| Multimodal score | – | 63.6 |
| Blended price $/M | $20.00 | $10.00 |
| Output speed | 58 tok/s | 38 tok/s |
| Time to first answer | 132.0 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 | 46.5% | 19.0% |
| Terminal-Bench | – | 79.8% |
| SimpleBench | – | 67.6% |
| SciCode | 60.1% | – |
| Cybench | – | 93.0% |
| Remote Labor Index | – | 4.2% |
| WeirdML | – | 77.9% |
| APEX-Agents | – | 32.4% |
| FrontierCode | – | 26.6% |
| GSO-Bench | – | 33.3% |
| Epoch Capabilities Index | – | 155.3 |
| LMArena Text | – | 1498 |
| LMArena Hard Prompts | – | 1527 |
| LMArena Coding | – | 1546 |
| LMArena WebDev | – | 1537 |
| LMArena Vision | – | 1311 |
| AA Intelligence Index | 53.2 | 31.9 |
| IFBench | – | 53.1% |
| AA-LCR | 83.0% | 78.0% |
| MMMU-Pro | – | 75.4% |
| AA-Omniscience | 42.4 | 13.7 |
| Terminal-Bench Hard | – | 48.5% |
| GPQA Diamond (AA) | 93.4% | 89.6% |
| Humanity's Last Exam (AA) | 58.7% | 39.9% |
| SciCode (AA) | 60.9% | – |
| τ²-Bench Telecom (AA) | – | 92.1% |
| HiL-Bench | – | 38.3% |
| EQ-Bench 4 | – | 1223 |
| Kagi LLM Benchmark | – | 72.4% |
| SWE-bench Verified (bash only) | – | 75.6% |
| SWE-bench Verified (any scaffold) | – | 75.6% |
| ARC-AGI-1 | 96.5% | – |
| ARC-AGI-2 | 90.0% | – |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.