Claude Opus 4.6 vs Muse Spark 1.1
Claude Opus 4.6 
Muse Spark 1.1 Verdict
- Muse Spark 1.1 leads on quality: 64.9 vs 63.1.
- Claude Opus 4.6 is stronger in agents & tools, long context, maths.
- Muse Spark 1.1 is stronger in coding, composite, human preference, instruction following, knowledge, multimodal, reasoning.
- Muse Spark 1.1 is 5.0× cheaper ($2.00 vs $10.00 per 1M blended).
- Muse Spark 1.1 streams 5.1× faster (194 vs 38 tokens per second).
| Metric | Claude Opus 4.6 | Muse Spark 1.1 |
|---|---|---|
| BenchLeader Index | 63.1 | 64.9 |
| Agents & tools score | 68.8 | 62.7 |
| Coding score | 62.3 | 68.2 |
| Composite score | 69.2 | 72.2 |
| Human preference score | 64.6 | 66.0 |
| Instruction following score | 53.7 | 77.8 |
| Knowledge score | 70.5 | 73.1 |
| Long context score | 65.5 | 65.3 |
| Maths score | 65.6 | 52.2 |
| Multimodal score | 63.6 | 64.9 |
| Reasoning score | 62.3 | 68.8 |
| Blended price $/M | $10.00 | $2.00 |
| Output speed | 38 tok/s | 194 tok/s |
| Time to first answer | 2.0 s | 2.0 s |
| Context window | 1M | 1.0M |
| GPQA Diamond | 90.5% | – |
| OTIS Mock AIME | 94.4% | – |
| SWE-bench Verified (Epoch) | 78.7% | – |
| SimpleQA Verified | – | 57.8% |
| Humanity's Last Exam | 19.0% | – |
| Terminal-Bench | 79.8% | – |
| SimpleBench | 67.6% | – |
| SciCode | – | 58.2% |
| Cybench | 93.0% | – |
| Remote Labor Index | 4.2% | – |
| WeirdML | 77.9% | – |
| APEX-Agents | 32.4% | 41.9% |
| FrontierCode | 26.6% | – |
| ProofBench | – | 39.0% |
| GSO-Bench | 33.3% | – |
| Epoch Capabilities Index | 155.3 | 154.6 |
| LMArena Text | 1498 | 1492 |
| LMArena Hard Prompts | 1527 | 1511 |
| LMArena Coding | 1546 | 1531 |
| LMArena WebDev | 1537 | 1541 |
| LMArena Vision | 1311 | 1293 |
| LMArena Agent | – | -2.6 |
| AA Intelligence Index | 31.9 | 34.3 |
| IFBench | 53.1% | – |
| AA-LCR | 78.0% | 77.7% |
| MMMU-Pro | 75.4% | – |
| AA-Omniscience | 13.7 | 28.1 |
| Terminal-Bench Hard | 48.5% | – |
| GPQA Diamond (AA) | 89.6% | 89.8% |
| Humanity's Last Exam (AA) | 39.9% | 46.2% |
| SciCode (AA) | – | 58.8% |
| τ²-Bench Telecom (AA) | 92.1% | – |
| SWE-Bench Pro | – | 61.5% |
| MCP Atlas | – | 88.1% |
| MultiChallenge | – | 75.3% |
| PRBench Finance | – | 55.0% |
| PRBench Legal | – | 57.0% |
| MultiNRC | – | 65.6% |
| HiL-Bench | 38.3% | – |
| EQ-Bench 4 | 1223 | 1260 |
| 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.