Model comparison
GPT-5.6 Luna vs Muse Spark
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 50.6 on the Noometry Index.
Last verified . 24 shared benchmarks.
MetaSummary
- They share 24 benchmarks with published results for both. GPT-5.6 Luna scores higher in 4 categories and Muse Spark in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Luna leads 77.7 to 47.8.
- The biggest single-benchmark swing is ProofBench: 60% for GPT-5.6 Luna and 17% for Muse Spark.
Side by side
| GPT-5.6 Luna | Muse Spark | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 54.6 | 50.6 |
| Released | 2026-07-09 | 2026-04-08 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.20 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 52 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Luna leads
GPT-5.6 Luna: 54.5 (#28), Muse Spark: 46.2 (#69)
| Benchmark | GPT-5.6 Luna | Muse Spark |
|---|---|---|
| SciCode | 53.6% | 51.5% |
| LMArena Coding | 1466 | 1481 |
| DeepSWE | 67.2% | — |
| FrontierCode | 39.8% | — |
| CursorBench | 35.9% | — |
| LMArena WebDev | 1519 | — |
| WeirdML | 60.9% | — |
| ALE-Bench | 1,667 | — |
Agentic & Tool Use Not comparable
GPT-5.6 Luna: 34.4 (#45), Muse Spark: —
| Benchmark | GPT-5.6 Luna | Muse Spark |
|---|---|---|
| APEX-Agents | 43% | — |
| BALROG | 45.6% | — |
| GDP.pdf | 22.7% | — |
| Vending-Bench 2 | 4,095 | — |
Reasoning GPT-5.6 Luna leads
GPT-5.6 Luna: 47.6 (#43), Muse Spark: 35.9 (#67)
| Benchmark | GPT-5.6 Luna | Muse Spark |
|---|---|---|
| CritPt | 20.6% | 11.3% |
| LMArena Hard Prompts | 1451 | 1474 |
| Epoch Capabilities Index | 156.39 | 152.04 |
| ARC-AGI-2 | 59.5% | — |
| SimpleBench | 46.8% | — |
| Kagi LLM Benchmark | 49.1% | — |
| NYT Connections (extended) | 69.4% | — |
| ARC-AGI-1 | 88% | — |
| Chess Puzzles | 40% | — |
| Mystery Game Puzzles | 21% | — |
| DTBench | 89.1% | — |
| LMCA | 48.5% | — |
| Surface Evolver Bench | 61.9% | — |
Math GPT-5.6 Luna leads
GPT-5.6 Luna: 77.7 (#14), Muse Spark: 47.8 (#66)
| Benchmark | GPT-5.6 Luna | Muse Spark |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.3% | 88.9% |
| ProofBench | 60% | 17% |
| LMArena Math | 1458 | 1455 |
| FrontierMath (Tiers 1-3) | 82.1% | — |
| FrontierMath Tier 4 | 61% | — |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Muse Spark leads
GPT-5.6 Luna: 58.5 (#34), Muse Spark: 65.7 (#13)
| Benchmark | GPT-5.6 Luna | Muse Spark |
|---|---|---|
| GPQA Diamond | 91.6% | 89.8% |
| LMArena Expert | 1478 | 1457 |
| Humanity's Last Exam | — | 40.6% |
| SimpleQA Verified | 41% | — |
Multimodal Too close to call
GPT-5.6 Luna: 42.7 (#28), Muse Spark: 43.4 (#24)
| Benchmark | GPT-5.6 Luna | Muse Spark |
|---|---|---|
| LMArena Vision | 1258 | 1306 |
| LMArena Document | 1457 | 1444 |
| Blueprint-Bench 2 | 22.6% | — |
| Furniture Assembly | 42.5% | — |
Multilingual Muse Spark leads
GPT-5.6 Luna: 52.8 (#78), Muse Spark: 56.1 (#24)
| Benchmark | GPT-5.6 Luna | Muse Spark |
|---|---|---|
| LMArena Non-English | 1417 | 1464 |
| LMArena Chinese | 1470 | 1509 |
| LMArena French | 1456 | 1497 |
| LMArena German | 1454 | 1497 |
| LMArena Korean | 1415 | 1459 |
| LMArena Russian | 1428 | 1466 |
| LMArena Spanish | 1448 | 1472 |
| LMArena Japanese | 1411 | — |
Instruction Following Too close to call
GPT-5.6 Luna: 75.6 (#57), Muse Spark: 75.9 (#51)
| Benchmark | GPT-5.6 Luna | Muse Spark |
|---|---|---|
| LMArena Instruction Following | 1437 | 1442 |
Long Context Too close to call
GPT-5.6 Luna: 43.9 (#82), Muse Spark: 44.4 (#69)
| Benchmark | GPT-5.6 Luna | Muse Spark |
|---|---|---|
| LMArena Longer Query | 1436 | 1451 |
Writing & Preference GPT-5.6 Luna leads
GPT-5.6 Luna: 68.0 (#29), Muse Spark: 66.0 (#39)
| Benchmark | GPT-5.6 Luna | Muse Spark |
|---|---|---|
| LMArena Text | 1431 | 1474 |
| LMArena Creative Writing | 1396 | 1459 |
| LMArena Multi-Turn | 1434 | 1477 |
| EQ-Bench Creative Writing | 1829 | — |
| EQ-Bench 4 | 1156 | — |
Frequently asked questions
Is GPT-5.6 Luna better than Muse Spark?
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 50.6 on the Noometry Index.
Is GPT-5.6 Luna or Muse Spark better for coding?
GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 46.2 in the Noometry coding category.
How many benchmarks do GPT-5.6 Luna and Muse Spark share?
24 benchmarks have published results for both models. GPT-5.6 Luna has 52 scored results on Noometry and Muse Spark has 27.