GPT-5.6 Terra vs Muse Spark 1.1
GPT-5.6 Terra 
Muse Spark 1.1 Verdict
- GPT-5.6 Terra (xhigh) and Muse Spark 1.1 are level on quality (64.2 vs 64.9).
- GPT-5.6 Terra (xhigh) is stronger in agents & tools, composite, human preference, long context, maths.
- Muse Spark 1.1 is stronger in coding, instruction following, knowledge, multimodal, reasoning.
- Muse Spark 1.1 is 2.3× cheaper ($2.00 vs $4.50 per 1M blended).
- Muse Spark 1.1 streams 2.1× faster (194 vs 93 tokens per second).
| Metric | GPT-5.6 Terra (xhigh) | Muse Spark 1.1 |
|---|---|---|
| BenchLeader Index | 64.2 | 64.9 |
| Agents & tools score | 69.6 | 62.7 |
| Coding score | 63.7 | 68.2 |
| Composite score | 77.1 | 72.2 |
| Human preference score | 66.3 | 66.0 |
| Instruction following score | 65.0 | 77.8 |
| Knowledge score | 60.9 | 73.1 |
| Long context score | 66.0 | 65.3 |
| Maths score | 71.0 | 52.2 |
| Multimodal score | 63.0 | 64.9 |
| Reasoning score | 58.2 | 68.8 |
| Blended price $/M | $4.50 | $2.00 |
| Output speed | 93 tok/s | 194 tok/s |
| Time to first answer | 32.5 s | 2.0 s |
| Context window | 1M | 1.0M |
| SimpleQA Verified | – | 57.8% |
| SimpleBench | 48.9% | – |
| SciCode | 51.6% | 58.2% |
| APEX-Agents | – | 41.9% |
| ProofBench | 74.0% | 39.0% |
| Epoch Capabilities Index | – | 154.6 |
| LMArena Text | 1466 | 1492 |
| LMArena Hard Prompts | 1491 | 1511 |
| LMArena Coding | 1519 | 1531 |
| LMArena WebDev | 1521 | 1541 |
| LMArena Vision | 1269 | 1293 |
| LMArena Agent | 1.5 | -2.6 |
| AA Intelligence Index | 38.2 | 34.3 |
| IFBench | 66.3% | – |
| AA-LCR | 79.0% | 77.7% |
| MMMU-Pro | 79.5% | – |
| AA-Omniscience | -3.0 | 28.1 |
| Terminal-Bench Hard | 62.9% | – |
| GPQA Diamond (AA) | 90.8% | 89.8% |
| Humanity's Last Exam (AA) | 41.9% | 46.2% |
| SciCode (AA) | 52.3% | 58.8% |
| τ²-Bench Telecom (AA) | 80.4% | – |
| LiveCodeBench | 85.9% | – |
| MMLU-Pro | 86.7% | – |
| IOI | 65.3% | – |
| CorpFin | 65.3% | – |
| TaxEval | 76.2% | – |
| GPQA Diamond (Vals) | 90.9% | – |
| SWE-Bench Pro | – | 61.5% |
| MCP Atlas | – | 88.1% |
| MultiChallenge | – | 75.3% |
| PRBench Finance | – | 55.0% |
| PRBench Legal | – | 57.0% |
| MultiNRC | – | 65.6% |
| EQ-Bench 4 | – | 1260 |
| ARC-AGI-1 | 94.0% | – |
| ARC-AGI-2 | 74.2% | – |
| ARC-AGI-3 | 0.7% | – |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.