Up to four models
GPT-6 Luna vs Muse Spark 1.2
On the index (GPT-4 = 100), GPT-6 Luna scores 162 and Muse Spark 1.2 162. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
GPT-6 Luna
162146–178
WeightsClosed
ReleasedSep 2026
Price in / out$0.10 / $0.50
Context1.05M
Core tests taken6 of 9
162146–178
WeightsClosed
ReleasedAug 2026
Price in / out$1.25 / $4.25
Context1.05M
Core tests taken3 of 9
GPT-6 Luna and Muse Spark 1.2 are too close to call (0 points apart, within the uncertainty), ahead on 1 of 3 shared core tests, and costs 10× less.
Test by testthird-party results
GPT-6 Luna
Muse Spark 1.2
Every result as a table
| Test | GPT-6 Luna | Muse Spark 1.2 |
|---|---|---|
| Reasoning & knowledge | ||
| LiveBench Data Analysis | 73.4% | 76.5% |
| LiveBench IF | 55.9% | 74.3% |
| LiveBench Language | 73.8% | 78.6% |
| LiveBench Reasoning | 81.8% | 90.0% |
| SimpleQA Verified | 41.4% | 60.3% |
| Maths | ||
| LiveBench Mathematics | 89.1% | 91.2% |
| ProofBench v1.1 | 64.0% | 43.0% |
| Chess puzzles | 36.0% | 27.0% |
| Coding | ||
| Terminal-Bench 4.0 | 13.6% | 6.1% |
| LiveBench Agentic Coding | 51.2% | 57.6% |
| LiveBench Coding | 79.0% | 77.5% |
| Code Migration | 42.5% | 29.9% |
| CyberBench v1.1 | 76.3% | 69.5% |
| IOI | 55.6% | 21.8% |
| ProgramBench | 0.5% | 0.5% |
| Vibe Code Bench v1.1 | 81.7% | 79.1% |
| Agents | ||
| APEX-Agents | 44.3% | 45.4% |
| BioMysteryBench | 61.5% | 64.8% |
| SRE Bench | 2.7% | 0.0% |
| DTBench | 90.1% | 94.7% |
| LMCA | 44.5% | 48.4% |
| Professional work | ||
| Finance Agent (v2) | 49.9% | 60.6% |
| EMB | 68.5% | 57.0% |
| Legal Research Bench | 30.3% | 43.8% |
| MedCode | 44.7% | 49.4% |
| MedScribe | 83.7% | 90.1% |
| Public Benefits Bench v1.1 | 57.6% | 68.5% |
| Tax Agent Bench | 20.0% | 32.8% |
Tests at least two of these models have taken. Bold rows are the core tests.