GPT-5.6 Luna scores 46.2 on the Respan Index (#33), ahead of Muse Spark 1.1 at 45.1 (#34). Muse Spark 1.1 leads on 9 of the 17 benchmarks both report. GPT-5.6 Luna is 4.4x cheaper per token at list price. GPT-5.6 Luna has the larger context window (1.1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| LiveBench Data Analysis (LiveBench) |
| 78% |
| 72.5% |
| LiveBench Instruction Following (LiveBench) | 60.1% | 69.6% |
| LiveBench Language (LiveBench) | 72.6% | 74.3% |
| LiveBench Mathematics (LiveBench) | 87.2% | 87.1% |
| LiveBench Reasoning (LiveBench) | 85.6% | 87.7% |
| SAGE (Vals AI) | 44.2% | 45.6% |
| SimpleQA Verified | 41% | 57.8% |
| Vending-Bench 2 (Andon Labs) | 4094.71 | 6520.48 |
| LMArena Agent (LMArena) | -0.0117 | -0.0488 |
| LMArena Elo | 1453 | 1491 |
| LMArena Vision (LMArena) | 1260 | 1281 |
| DeepSWE v1.1 | 67.2% | 53.3% |
| SWE-Bench Pro | 62.7% | 61.5% |
Higher is better. Each score links to where it was published.