Claude Fable 5 vs Grok 4.5
Claude Fable 5 
Grok 4.5 Verdict
- Claude Fable 5 leads on quality: 70.4 vs 63.4.
- Claude Fable 5 is stronger in agents & tools, coding, composite, human preference, instruction following, long context, multimodal, reasoning.
- Grok 4.5 is stronger in knowledge.
- Grok 4.5 is 6.7× cheaper ($3.00 vs $20.00 per 1M blended).
| Metric | Claude Fable 5 | Grok 4.5 |
|---|---|---|
| BenchLeader Index | 70.4 | 63.4 |
| Agents & tools score | 74.4 | 58.6 |
| Coding score | 70.6 | 59.9 |
| Composite score | 91.6 | 66.0 |
| Human preference score | 71.3 | 66.9 |
| Instruction following score | 62.6 | – |
| Knowledge score | 71.3 | 76.0 |
| Long context score | 67.8 | 66.2 |
| Multimodal score | 69.6 | 64.8 |
| Reasoning score | 75.6 | 68.9 |
| Blended price $/M | $20.00 | $3.00 |
| Output speed | 63 tok/s | 56 tok/s |
| Time to first answer | 92.2 s | 12.7 s |
| Context window | 1M | 500k |
| SimpleBench | – | 70.0% |
| Remote Labor Index | 16.1% | – |
| WeirdML | – | 46.4% |
| APEX-Agents | 45.0% | 34.2% |
| FrontierCode | 53.5% | 42.4% |
| Epoch Capabilities Index | 163.4 | 153.9 |
| LMArena Text | 1507 | 1471 |
| LMArena Hard Prompts | 1533 | 1495 |
| LMArena Coding | 1551 | 1523 |
| LMArena WebDev | 1628 | 1556 |
| LMArena Vision | 1330 | 1291 |
| LMArena Agent | – | 3.9 |
| LiveBench | – | 75.8% |
| LiveBench Reasoning | – | 87.2% |
| LiveBench Coding | – | 68.6% |
| LiveBench Agentic Coding | – | 56.5% |
| LiveBench Mathematics | – | 90.8% |
| LiveBench Data Analysis | – | 73.0% |
| LiveBench Language | – | 82.8% |
| AA Intelligence Index | 49.7 | 39.1 |
| IFBench | 63.5% | – |
| AA-LCR | 82.3% | 79.3% |
| MMMU-Pro | – | 80.4% |
| AA-Omniscience | 43.3 | 25.3 |
| Terminal-Bench Hard | 62.9% | – |
| GPQA Diamond (AA) | 92.6% | 93.1% |
| Humanity's Last Exam (AA) | 55.5% | 42.7% |
| SciCode (AA) | 61.0% | 55.0% |
| τ²-Bench Telecom (AA) | 98.5% | – |
| 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% | – |
| EQ-Bench 4 | 1340 | – |
| Kagi LLM Benchmark | 88.8% | 83.5% |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.