Model comparison
Muse Spark vs Step 3.7 Flash
Muse Spark is the stronger model overall, scoring 50.6 to 37.3 on the Noometry Index.
Last verified . 2 shared benchmarks.
Meta
StepFunSummary
- They share 2 benchmarks with published results for both. Muse Spark scores higher in 3 categories and Step 3.7 Flash in 0 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Muse Spark leads 35.9 to 21.6.
- The biggest single-benchmark swing is SciCode: 51.5% for Muse Spark and 40% for Step 3.7 Flash.
- Step 3.7 Flash has downloadable open weights; the other is API-only.
Side by side
| Muse Spark | Step 3.7 Flash | |
|---|---|---|
| Provider | Meta | StepFun |
| Noometry Index | 50.6 | 37.3 |
| Released | 2026-04-08 | 2026-05-29 |
| Weights | Proprietary | Open |
| Context window | — | 256K |
| Max output | — | 256K |
| Input $ / M tokens | — | $0.18 |
| Output $ / M tokens | — | $1.11 |
| Results tracked | 27 | 5 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Muse Spark leads
Muse Spark: 46.2 (#69), Step 3.7 Flash: 40.0 (#150)
| Benchmark | Muse Spark | Step 3.7 Flash |
|---|---|---|
| SciCode | 51.5% | 40% |
| LMArena Coding | 1481 | — |
| ALE-Bench | — | 694.12 |
Reasoning Muse Spark leads
Muse Spark: 35.9 (#67), Step 3.7 Flash: 21.6 (#219)
| Benchmark | Muse Spark | Step 3.7 Flash |
|---|---|---|
| CritPt | 11.3% | 2.3% |
| NYT Connections (extended) | — | 39.7% |
| LMArena Hard Prompts | 1474 | — |
| Epoch Capabilities Index | 152.04 | — |
Math Muse Spark leads
Muse Spark: 47.8 (#66), Step 3.7 Flash: 42.9 (#82)
| Benchmark | Muse Spark | Step 3.7 Flash |
|---|---|---|
| MathArena Final-Answer Competitions | — | 68.5% |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 17% | — |
| LMArena Math | 1455 | — |
| FrontierMath (Feb 2025 set) | 39% | — |
| FrontierMath Tier 4 (v1) | 14.6% | — |
Knowledge Not comparable
Muse Spark: 65.7 (#13), Step 3.7 Flash: —
| Benchmark | Muse Spark | Step 3.7 Flash |
|---|---|---|
| GPQA Diamond | 89.8% | — |
| Humanity's Last Exam | 40.6% | — |
| LMArena Expert | 1457 | — |
Multimodal Not comparable
Muse Spark: 43.4 (#24), Step 3.7 Flash: —
| Benchmark | Muse Spark | Step 3.7 Flash |
|---|---|---|
| LMArena Vision | 1306 | — |
| LMArena Document | 1444 | — |
Multilingual Not comparable
Muse Spark: 56.1 (#24), Step 3.7 Flash: —
| Benchmark | Muse Spark | Step 3.7 Flash |
|---|---|---|
| LMArena Non-English | 1464 | — |
| LMArena Chinese | 1509 | — |
| LMArena French | 1497 | — |
| LMArena German | 1497 | — |
| LMArena Korean | 1459 | — |
| LMArena Russian | 1466 | — |
| LMArena Spanish | 1472 | — |
Instruction Following Not comparable
Muse Spark: 75.9 (#51), Step 3.7 Flash: —
| Benchmark | Muse Spark | Step 3.7 Flash |
|---|---|---|
| LMArena Instruction Following | 1442 | — |
Long Context Not comparable
Muse Spark: 44.4 (#69), Step 3.7 Flash: —
| Benchmark | Muse Spark | Step 3.7 Flash |
|---|---|---|
| LMArena Longer Query | 1451 | — |
Writing & Preference Not comparable
Muse Spark: 66.0 (#39), Step 3.7 Flash: —
| Benchmark | Muse Spark | Step 3.7 Flash |
|---|---|---|
| LMArena Text | 1474 | — |
| LMArena Creative Writing | 1459 | — |
| LMArena Multi-Turn | 1477 | — |
Frequently asked questions
Is Muse Spark better than Step 3.7 Flash?
Muse Spark is the stronger model overall, scoring 50.6 to 37.3 on the Noometry Index.
Is Muse Spark or Step 3.7 Flash better for coding?
Muse Spark scores higher on coding benchmarks: 46.2 versus 40.0 in the Noometry coding category.
How many benchmarks do Muse Spark and Step 3.7 Flash share?
2 benchmarks have published results for both models. Muse Spark has 27 scored results on Noometry and Step 3.7 Flash has 5.