Claude Opus 4.6 vs Muse Spark
Claude Opus 4.6 
Muse Spark Verdict
- Muse Spark leads on quality: 65.4 vs 63.1.
- Claude Opus 4.6 is stronger in agents & tools, composite, knowledge, maths.
- Muse Spark is stronger in coding, human preference, instruction following, multimodal, reasoning.
| Metric | Claude Opus 4.6 | Muse Spark |
|---|---|---|
| BenchLeader Index | 63.1 | 65.4 |
| Agents & tools score | 68.8 | 66.3 |
| Coding score | 62.3 | 66.0 |
| Composite score | 69.2 | 68.4 |
| Human preference score | 64.6 | 69.0 |
| Instruction following score | 53.7 | 79.6 |
| Knowledge score | 70.5 | 64.5 |
| Long context score | 65.5 | 65.5 |
| Maths score | 65.6 | 56.3 |
| Multimodal score | 63.6 | 65.9 |
| Reasoning score | 62.3 | 70.0 |
| Blended price $/M | $10.00 | – |
| Output speed | 38 tok/s | – |
| Time to first answer | 2.0 s | – |
| Context window | 1M | 262k |
| GPQA Diamond | 90.5% | 89.8% |
| OTIS Mock AIME | 94.4% | 88.9% |
| SWE-bench Verified (Epoch) | 78.7% | – |
| Humanity's Last Exam | 19.0% | 40.6% |
| Terminal-Bench | 79.8% | – |
| SimpleBench | 67.6% | – |
| SciCode | – | 51.5% |
| Cybench | 93.0% | – |
| Remote Labor Index | 4.2% | – |
| WeirdML | 77.9% | – |
| APEX-Agents | 32.4% | – |
| FrontierCode | 26.6% | – |
| ProofBench | – | 17.0% |
| GSO-Bench | 33.3% | – |
| Epoch Capabilities Index | 155.3 | 152.1 |
| LMArena Text | 1498 | 1488 |
| LMArena Hard Prompts | 1527 | 1505 |
| LMArena Coding | 1546 | 1526 |
| LMArena WebDev | 1537 | – |
| LMArena Vision | 1311 | 1306 |
| AA Intelligence Index | 31.9 | 31.3 |
| IFBench | 53.1% | 75.9% |
| AA-LCR | 78.0% | 78.0% |
| MMMU-Pro | 75.4% | 80.5% |
| AA-Omniscience | 13.7 | 7.2 |
| Terminal-Bench Hard | 48.5% | 45.5% |
| GPQA Diamond (AA) | 89.6% | 88.4% |
| Humanity's Last Exam (AA) | 39.9% | 40.7% |
| τ²-Bench Telecom (AA) | 92.1% | 91.5% |
| AIME (Vals) | – | 96.9% |
| MMLU-Pro | – | 87.3% |
| LegalBench | – | 84.2% |
| CorpFin | – | 65.1% |
| TaxEval | – | 77.7% |
| SWE-bench (Vals) | – | 74.4% |
| GPQA Diamond (Vals) | – | 89.7% |
| SWE-Bench Pro | – | 55.0% |
| MCP Atlas | – | 82.2% |
| MultiChallenge | – | 75.5% |
| PRBench Finance | – | 52.4% |
| PRBench Legal | – | 52.3% |
| MultiNRC | – | 59.0% |
| HiL-Bench | 38.3% | – |
| TutorBench | – | 68.5% |
| EQ-Bench 4 | 1223 | – |
| Kagi LLM Benchmark | 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.