GPT-5.4 Pro vs Muse Spark 1.3
GPT-5.4 Pro 
Muse Spark 1.3 Verdict
- GPT-5.4 Pro and Muse Spark 1.3 (xhigh) are level on quality (64.2 vs 63.9).
- GPT-5.4 Pro is stronger in instruction following, multimodal, reasoning.
- Muse Spark 1.3 (xhigh) is stronger in knowledge, agents & tools, coding, composite, long context.
- Muse Spark 1.3 (xhigh) is 34× cheaper ($2.00 vs $67.50 per 1M blended).
- Muse Spark 1.3 (xhigh) streams 193.4× faster (193 vs 1 tokens per second).
| Metric | GPT-5.4 Pro | Muse Spark 1.3 (xhigh) |
|---|---|---|
| BenchLeader Index | 64.2 | 63.9 |
| Instruction following score | 67.3 | – |
| Knowledge score | 73.2 | 75.0 |
| Multimodal score | 69.8 | 66.6 |
| Reasoning score | 75.1 | – |
| Agents & tools score | – | 60.4 |
| Coding score | – | 70.8 |
| Composite score | – | 78.6 |
| Long context score | – | 68.1 |
| Blended price $/M | $67.50 | $2.00 |
| Output speed | 1 tok/s | 193 tok/s |
| Time to first answer | 6.7 s | 40.0 s |
| Context window | 1.1M | 1M |
| Humanity's Last Exam | 44.3% | – |
| SimpleBench | 74.1% | – |
| Epoch Capabilities Index | 158.9 | – |
| 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 | – | 45.2 |
| AA-LCR | – | 83.0% |
| MMMU-Pro | – | 82.0% |
| AA-Omniscience | – | 23.1 |
| GPQA Diamond (AA) | – | 94.1% |
| Humanity's Last Exam (AA) | – | 47.5% |
| SciCode (AA) | – | 59.7% |
| Terminal-Bench 2.1 (Vals) | – | 72.3% |
| Vals Index | – | 60.3 |
| MultiChallenge | 69.2% | – |
| VISTA | 53.9% | – |
| MultiNRC | 62.3% | – |
| TutorBench | 56.6% | – |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.