Muse Spark 1.1 scores 45.1 on the Respan Index (#34), ahead of GPT-5.2 at 43.5 (#36). Muse Spark 1.1 leads on 11 of the 15 benchmarks both report. Muse Spark 1.1 is 2.4x cheaper per token at list price. Muse Spark 1.1 has the larger context window (1.0M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| LiveBench Language (LiveBench) |
| 79.8% |
| 74.3% |
| LiveBench Mathematics (LiveBench) | 93.2% | 87.1% |
| LiveBench Reasoning (LiveBench) | 83.2% | 87.7% |
| SAGE (Vals AI) | 49.3% | 45.6% |
| SimpleQA Verified | 37.1% | 57.8% |
| Vending-Bench 2 (Andon Labs) | 3591.33 | 6520.48 |
| LMArena Elo | 1412 | 1491 |
| LMArena Vision (LMArena) | 1243 | 1281 |
| LMArena WebDev (LMArena) | 1416 | 1542 |
| tau2-bench Banking Knowledge (Sierra) | 32.2% | 40.5% |
| SWE-Bench Pro | 55.6% | 61.5% |
Higher is better. Each score links to where it was published.