Model comparison
GPT-6 Luna vs Muse Spark
GPT-6 Luna is the stronger model overall, scoring 53.3 to 50.6 on the Noometry Index.
Last verified . 23 shared benchmarks.
MetaSummary
- They share 23 benchmarks with published results for both. GPT-6 Luna scores higher in 3 categories and Muse Spark in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 47.8.
- The biggest single-benchmark swing is ProofBench: 64% for GPT-6 Luna and 17% for Muse Spark.
Side by side
| GPT-6 Luna | Muse Spark | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 53.3 | 50.6 |
| Released | 2026-09-22 | 2026-04-08 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.10 | — |
| Output $ / M tokens | $0.50 | — |
| Results tracked | 42 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
GPT-6 Luna: 55.5 (#25), Muse Spark: 46.2 (#69)
| Benchmark | GPT-6 Luna | Muse Spark |
|---|---|---|
| SciCode | 54.6% | 51.5% |
| LMArena Coding | 1439 | 1481 |
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| LMArena WebDev | 1581 | — |
| ALE-Bench | 1,577 | — |
Agentic & Tool Use Not comparable
GPT-6 Luna: 33.3 (#54), Muse Spark: —
| Benchmark | GPT-6 Luna | Muse Spark |
|---|---|---|
| APEX-Agents | 44.3% | — |
| GDP.pdf | 23% | — |
Reasoning GPT-6 Luna leads
GPT-6 Luna: 48.2 (#41), Muse Spark: 35.9 (#67)
| Benchmark | GPT-6 Luna | Muse Spark |
|---|---|---|
| CritPt | 19.4% | 11.3% |
| LMArena Hard Prompts | 1411 | 1474 |
| Epoch Capabilities Index | 156.28 | 152.04 |
| ARC-AGI-2 | 59.3% | — |
| NYT Connections (extended) | 68.7% | — |
| ARC-AGI-1 | 86.7% | — |
| Chess Puzzles | 31% | — |
| Mystery Game Puzzles | 7% | — |
| DTBench | 90.1% | — |
| LMCA | 44.5% | — |
Math GPT-6 Luna leads
GPT-6 Luna: 76.1 (#15), Muse Spark: 47.8 (#66)
| Benchmark | GPT-6 Luna | Muse Spark |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 88.9% |
| ProofBench | 64% | 17% |
| LMArena Math | 1416 | 1455 |
| FrontierMath (Tiers 1-3) | 78.9% | — |
| FrontierMath Tier 4 | 56.1% | — |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Muse Spark leads
GPT-6 Luna: 57.0 (#41), Muse Spark: 65.7 (#13)
| Benchmark | GPT-6 Luna | Muse Spark |
|---|---|---|
| GPQA Diamond | 90.5% | 89.8% |
| LMArena Expert | 1444 | 1457 |
| Humanity's Last Exam | — | 40.6% |
| SimpleQA Verified | 41.4% | — |
Multimodal Too close to call
GPT-6 Luna: 42.4 (#30), Muse Spark: 43.4 (#24)
| Benchmark | GPT-6 Luna | Muse Spark |
|---|---|---|
| LMArena Vision | 1217 | 1306 |
| Blueprint-Bench 2 | 31.2% | — |
| Furniture Assembly | 44.2% | — |
| LMArena Document | — | 1444 |
Multilingual Muse Spark leads
GPT-6 Luna: 50.5 (#117), Muse Spark: 56.1 (#24)
| Benchmark | GPT-6 Luna | Muse Spark |
|---|---|---|
| LMArena Non-English | 1386 | 1464 |
| LMArena Chinese | 1433 | 1509 |
| LMArena French | 1420 | 1497 |
| LMArena German | 1369 | 1497 |
| LMArena Korean | 1360 | 1459 |
| LMArena Russian | 1394 | 1466 |
| LMArena Spanish | 1393 | 1472 |
| LMArena Japanese | 1369 | — |
Instruction Following Muse Spark leads
GPT-6 Luna: 74.3 (#99), Muse Spark: 75.9 (#51)
| Benchmark | GPT-6 Luna | Muse Spark |
|---|---|---|
| LMArena Instruction Following | 1409 | 1442 |
Long Context Muse Spark leads
GPT-6 Luna: 43.0 (#111), Muse Spark: 44.4 (#69)
| Benchmark | GPT-6 Luna | Muse Spark |
|---|---|---|
| LMArena Longer Query | 1409 | 1451 |
Writing & Preference Muse Spark leads
GPT-6 Luna: 58.3 (#119), Muse Spark: 66.0 (#39)
| Benchmark | GPT-6 Luna | Muse Spark |
|---|---|---|
| LMArena Text | 1391 | 1474 |
| LMArena Creative Writing | 1363 | 1459 |
| LMArena Multi-Turn | 1396 | 1477 |
Frequently asked questions
Is GPT-6 Luna better than Muse Spark?
GPT-6 Luna is the stronger model overall, scoring 53.3 to 50.6 on the Noometry Index.
Is GPT-6 Luna or Muse Spark better for coding?
GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 46.2 in the Noometry coding category.
How many benchmarks do GPT-6 Luna and Muse Spark share?
23 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Muse Spark has 27.