GPT-5.6 Sol vs Muse Spark 1.3
GPT-5.6 Sol 
Muse Spark 1.3 Verdict
- GPT-5.6 Sol (high) leads on quality: 68.0 vs 63.9.
- GPT-5.6 Sol (high) is stronger in agents & tools, coding, composite, instruction following, reasoning.
- Muse Spark 1.3 (xhigh) is stronger in knowledge, long context, multimodal.
- Muse Spark 1.3 (xhigh) is 4.0× cheaper ($2.00 vs $8.00 per 1M blended).
- Muse Spark 1.3 (xhigh) streams 3.3× faster (193 vs 59 tokens per second).
| Metric | GPT-5.6 Sol (high) | Muse Spark 1.3 (xhigh) |
|---|---|---|
| BenchLeader Index | 68.0 | 63.9 |
| Agents & tools score | 87.6 | 60.4 |
| Coding score | 72.0 | 70.8 |
| Composite score | 82.5 | 78.6 |
| Instruction following score | 67.6 | – |
| Knowledge score | 73.7 | 75.0 |
| Long context score | 67.4 | 68.1 |
| Multimodal score | 66.5 | 66.6 |
| Reasoning score | 62.9 | – |
| Blended price $/M | $8.00 | $2.00 |
| Output speed | 59 tok/s | 193 tok/s |
| Time to first answer | 31.4 s | 40.0 s |
| Context window | 1M | 1M |
| SciCode | 56.9% | – |
| WeirdML | 88.8% | – |
| LMArena WebDev | – | 1625 |
| LiveBench | – | 81.6% |
| LiveBench Reasoning | – | 89.7% |
| LiveBench Coding | – | 81.1% |
| LiveBench Agentic Coding | – | 64.1% |
| LiveBench Mathematics | – | 96.0% |
| LiveBench Data Analysis | – | 79.6% |
| LiveBench Language | – | 82.8% |
| AA Intelligence Index | 42.5 | 45.2 |
| IFBench | 69.2% | – |
| AA-LCR | 81.7% | 83.0% |
| MMMU-Pro | 81.8% | 82.0% |
| AA-Omniscience | 20.4 | 23.1 |
| Terminal-Bench Hard | 62.1% | – |
| GPQA Diamond (AA) | 92.8% | 94.1% |
| Humanity's Last Exam (AA) | 46.0% | 47.5% |
| SciCode (AA) | 57.8% | 59.7% |
| τ²-Bench Telecom (AA) | 83.3% | – |
| Terminal-Bench 2.1 (Vals) | – | 72.3% |
| Vals Index | – | 60.3 |
| EnigmaEval | 37.1% | – |
| ARC-AGI-1 | 97.0% | – |
| ARC-AGI-2 | 85.4% | – |
| ARC-AGI-3 | 2.2% | – |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.