Muse Spark 1.2 vs Muse Spark 1.3
Muse Spark 1.2 Verdict
- Muse Spark 1.3 (xhigh) leads on quality: 63.9 vs 62.6.
- Muse Spark 1.2 is stronger in composite, knowledge, maths, reasoning.
- Muse Spark 1.3 (xhigh) is stronger in coding, long context, agents & tools, multimodal.
- They cost about the same ($2.00 per 1M blended).
| Metric | Muse Spark 1.2 | Muse Spark 1.3 (xhigh) |
|---|---|---|
| BenchLeader Index | 62.6 | 63.9 |
| Coding score | 66.5 | 70.8 |
| Composite score | 79.1 | 78.6 |
| Knowledge score | 76.9 | 75.0 |
| Long context score | 66.0 | 68.1 |
| Maths score | 54.3 | – |
| Reasoning score | 70.2 | – |
| Agents & tools score | – | 60.4 |
| Multimodal score | – | 66.6 |
| Blended price $/M | $2.00 | $2.00 |
| Output speed | 219 tok/s | 193 tok/s |
| Time to first answer | 23.7 s | 40.0 s |
| Context window | 1.0M | 1M |
| SimpleBench | 74.5% | – |
| ProofBench | 43.0% | – |
| Epoch Capabilities Index | 155.5 | – |
| 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 | 39.8 | 45.2 |
| AA-LCR | 79.0% | 83.0% |
| MMMU-Pro | – | 82.0% |
| AA-Omniscience | 27.2 | 23.1 |
| GPQA Diamond (AA) | 90.4% | 94.1% |
| Humanity's Last Exam (AA) | 45.5% | 47.5% |
| SciCode (AA) | 57.4% | 59.7% |
| IOI | 49.5% | – |
| Terminal-Bench 2.1 (Vals) | – | 72.3% |
| Vals Index | – | 60.3 |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.