Model comparison
Llama 4 Scout vs Muse Spark 1.2
Muse Spark 1.2 is the stronger model overall, scoring 50.3 to 27.7 on the Noometry Index. Llama 4 Scout costs 13× less per token, which makes it the better buy when Muse Spark 1.2's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
MetaSummary
- They share 21 benchmarks with published results for both. Llama 4 Scout scores higher in 0 categories and Muse Spark 1.2 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Muse Spark 1.2 leads 51.3 to 9.1.
- The biggest single-benchmark swing is SciCode: 17% for Llama 4 Scout and 56.4% for Muse Spark 1.2.
- Llama 4 Scout is cheaper at $0.10 / $0.30 per million input/output tokens, against $1.25 / $4.25 for Muse Spark 1.2.
- Muse Spark 1.2 accepts more context: 1.05M tokens versus 128K.
- Llama 4 Scout has downloadable open weights; the other is API-only.
Side by side
| Llama 4 Scout | Muse Spark 1.2 | |
|---|---|---|
| Provider | Meta | Meta |
| Noometry Index | 27.7 | 50.3 |
| Released | 2025-04-05 | 2026-08-05 |
| Weights | Open | Proprietary |
| Context window | 128K | 1.05M |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.10 | $1.25 |
| Output $ / M tokens | $0.30 | $4.25 |
| Results tracked | 43 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Muse Spark 1.2 leads
Llama 4 Scout: 20.2 (#339), Muse Spark 1.2: 49.2 (#51)
| Benchmark | Llama 4 Scout | Muse Spark 1.2 |
|---|---|---|
| SciCode | 17% | 56.4% |
| LMArena Coding | 1286 | 1495 |
| DeepSWE | — | 54.9% |
| SWE-bench Verified (bash only) | 9.1% | — |
| LMArena WebDev | — | 1533 |
| FrontierSWE | — | 12% |
| WeirdML | — | 60.3% |
| BigCodeBench Complete | 43.1% | — |
Agentic & Tool Use Muse Spark 1.2 leads
Llama 4 Scout: 24.6 (#119), Muse Spark 1.2: 29.4 (#87)
| Benchmark | Llama 4 Scout | Muse Spark 1.2 |
|---|---|---|
| APEX-Agents | — | 36.4% |
| Berkeley Function Calling Leaderboard | 28.1% | — |
| GDP.pdf | — | 16% |
Reasoning Muse Spark 1.2 leads
Llama 4 Scout: 9.1 (#345), Muse Spark 1.2: 51.3 (#34)
| Benchmark | Llama 4 Scout | Muse Spark 1.2 |
|---|---|---|
| CritPt | 0% | 17.7% |
| LMArena Hard Prompts | 1266 | 1486 |
| DTBench | 57.9% | 94.7% |
| LMCA | 12% | 48.4% |
| Epoch Capabilities Index | 129.64 | 154.87 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | — | 74.5% |
| Kagi LLM Benchmark | 36.9% | — |
| NYT Connections (extended) | — | 79.2% |
| ARC-AGI-1 | 0.5% | — |
| ForecastBench | 57.5 | — |
Math Muse Spark 1.2 leads
Llama 4 Scout: 19.6 (#286), Muse Spark 1.2: 46.4 (#70)
| Benchmark | Llama 4 Scout | Muse Spark 1.2 |
|---|---|---|
| LMArena Math | 1287 | 1471 |
| OTIS Mock AIME 2024-2025 | 7.8% | — |
| ProofBench | — | 43% |
| Omni-MATH | 37.3% | — |
| MATH Level 5 | 62.3% | — |
| FrontierMath (Feb 2025 set) | 0% | — |
Knowledge Muse Spark 1.2 leads
Llama 4 Scout: 31.9 (#217), Muse Spark 1.2: 54.1 (#53)
| Benchmark | Llama 4 Scout | Muse Spark 1.2 |
|---|---|---|
| LMArena Expert | 1235 | 1480 |
| GPQA Diamond | 51.8% | — |
| SimpleQA Verified | — | 60.3% |
| MMLU-Pro | 74.2% | — |
| Vectara Hallucination Rate | 7.7% | — |
| GPQA (HELM) | 50.7% | — |
Multimodal Muse Spark 1.2 leads
Llama 4 Scout: 32.2 (#102), Muse Spark 1.2: 43.4 (#25)
| Benchmark | Llama 4 Scout | Muse Spark 1.2 |
|---|---|---|
| LMArena Vision | 1118 | 1305 |
| SpatialViz-Bench | 34.2% | — |
Multilingual Muse Spark 1.2 leads
Llama 4 Scout: 41.0 (#212), Muse Spark 1.2: 57.1 (#11)
| Benchmark | Llama 4 Scout | Muse Spark 1.2 |
|---|---|---|
| LMArena Non-English | 1252 | 1478 |
| LMArena Chinese | 1255 | 1511 |
| LMArena French | 1282 | 1513 |
| LMArena Russian | 1263 | 1487 |
| LMArena Spanish | 1278 | 1498 |
| LMArena German | 1272 | — |
| LMArena Japanese | 1206 | — |
| LMArena Korean | 1207 | — |
Instruction Following Muse Spark 1.2 leads
Llama 4 Scout: 65.8 (#217), Muse Spark 1.2: 76.7 (#36)
| Benchmark | Llama 4 Scout | Muse Spark 1.2 |
|---|---|---|
| LMArena Instruction Following | 1248 | 1461 |
| IFEval | 81.8% | — |
Long Context Muse Spark 1.2 leads
Llama 4 Scout: 27.5 (#294), Muse Spark 1.2: 45.2 (#48)
| Benchmark | Llama 4 Scout | Muse Spark 1.2 |
|---|---|---|
| LMArena Longer Query | 1265 | 1475 |
| Fiction.LiveBench | 36% | — |
Writing & Preference Muse Spark 1.2 leads
Llama 4 Scout: 37.0 (#261), Muse Spark 1.2: 72.3 (#14)
| Benchmark | Llama 4 Scout | Muse Spark 1.2 |
|---|---|---|
| LMArena Text | 1279 | 1482 |
| LMArena Creative Writing | 1249 | 1449 |
| EQ-Bench Creative Writing | 783 | 1840 |
| LMArena Multi-Turn | 1280 | 1494 |
| WildBench | 78% | — |
Frequently asked questions
Is Llama 4 Scout better than Muse Spark 1.2?
Muse Spark 1.2 is the stronger model overall, scoring 50.3 to 27.7 on the Noometry Index. Llama 4 Scout costs 13× less per token, which makes it the better buy when Muse Spark 1.2's lead doesn't matter for your workload.
Which is cheaper, Llama 4 Scout or Muse Spark 1.2?
Llama 4 Scout is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Muse Spark 1.2 lists at $1.25 and $4.25.
Is Llama 4 Scout or Muse Spark 1.2 better for coding?
Muse Spark 1.2 scores higher on coding benchmarks: 49.2 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
Muse Spark 1.2 does, with 1.05M tokens against 128K.
How many benchmarks do Llama 4 Scout and Muse Spark 1.2 share?
21 benchmarks have published results for both models. Llama 4 Scout has 43 scored results on Noometry and Muse Spark 1.2 has 31.