GPT-6 Astra vs Muse Spark 1.2
GPT-6 Astra 
Muse Spark 1.2 Verdict
- GPT-6 Astra (max) leads on quality: 70.6 vs 62.6.
- GPT-6 Astra (max) is stronger in agents & tools, coding, knowledge, maths, reasoning.
- Muse Spark 1.2 is stronger in composite, long context.
- Muse Spark 1.2 is 11× cheaper ($2.00 vs $22.00 per 1M blended).
- Muse Spark 1.2 streams 4.1× faster (219 vs 54 tokens per second).
| Metric | GPT-6 Astra (max) | Muse Spark 1.2 |
|---|---|---|
| BenchLeader Index | 70.6 | 62.6 |
| Agents & tools score | 68.2 | – |
| Coding score | 79.0 | 66.5 |
| Composite score | 73.1 | 79.1 |
| Knowledge score | 79.8 | 76.9 |
| Maths score | 77.0 | 54.3 |
| Reasoning score | 70.9 | 70.2 |
| Long context score | – | 66.0 |
| Blended price $/M | $22.00 | $2.00 |
| Output speed | 54 tok/s | 219 tok/s |
| Time to first answer | 328.6 s | 23.7 s |
| Context window | 1.1M | 1.0M |
| GPQA Diamond | 95.8% | – |
| FrontierMath Tiers 1–3 | 93.7% | – |
| FrontierMath Tier 4 | 97.6% | – |
| OTIS Mock AIME | 100.0% | – |
| SimpleQA Verified | 75.6% | – |
| Terminal-Bench | 58.2% | – |
| SimpleBench | – | 74.5% |
| SciCode | 56.5% | – |
| FrontierCode | 53.3% | – |
| ProofBench | – | 43.0% |
| Epoch Capabilities Index | – | 155.5 |
| LMArena WebDev | 1796 | – |
| LMArena Agent | 12.5 | – |
| LiveBench | 82.2% | – |
| LiveBench Reasoning | 92.7% | – |
| LiveBench Coding | 80.4% | – |
| LiveBench Agentic Coding | 57.3% | – |
| LiveBench Mathematics | 96.8% | – |
| LiveBench Data Analysis | 83.0% | – |
| LiveBench Language | 89.4% | – |
| AA Intelligence Index | – | 39.8 |
| AA-LCR | – | 79.0% |
| AA-Omniscience | – | 27.2 |
| GPQA Diamond (AA) | – | 90.4% |
| Humanity's Last Exam (AA) | – | 45.5% |
| SciCode (AA) | – | 57.4% |
| IOI | – | 49.5% |
| Terminal-Bench 2.1 (Vals) | 87.3% | – |
| Vals Index | 66.6 | – |
| ARC-AGI-1 | 97.5% | – |
| ARC-AGI-2 | 95.0% | – |
| ARC-AGI-3 | 62.7% | – |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.