Model comparison
Muse Spark 1.2 vs Qwen2.5 72B Instruct
Muse Spark 1.2 is the stronger model overall, scoring 49.8 to 31.5 on the Noometry Index.
Last verified . 15 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 15 benchmarks with published results for both. Muse Spark 1.2 scores higher in 9 categories and Qwen2.5 72B Instruct in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Muse Spark 1.2 leads 50.4 to 21.4.
- The biggest single-benchmark swing is WeirdML: 60.3% for Muse Spark 1.2 and 16% for Qwen2.5 72B Instruct.
- Muse Spark 1.2 is cheaper at $1.25 / $4.25 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- Muse Spark 1.2 accepts more context: 1.05M tokens versus 131K.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| Muse Spark 1.2 | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 49.8 | 31.5 |
| Released | 2026-08-05 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 131K | 8K |
| Input $ / M tokens | $1.25 | $1.40 |
| Output $ / M tokens | $4.25 | $5.60 |
| Results tracked | 28 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Muse Spark 1.2 leads
Muse Spark 1.2: 48.6 (#49), Qwen2.5 72B Instruct: 32.8 (#256)
| Benchmark | Muse Spark 1.2 | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 60.3% | 16% |
| LMArena Coding | 1497 | 1293 |
| DeepSWE | 54.9% | — |
| LMArena WebDev | 1532 | — |
| FrontierSWE | 12% | — |
| SciCode | 56.4% | — |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Muse Spark 1.2 leads
Muse Spark 1.2: 29.5 (#77), Qwen2.5 72B Instruct: 22.0 (#124)
| Benchmark | Muse Spark 1.2 | Qwen2.5 72B Instruct |
|---|---|---|
| APEX-Agents | 36.4% | — |
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| GDP.pdf | 16% | — |
| METR Time Horizons | — | 35.8% |
Reasoning Muse Spark 1.2 leads
Muse Spark 1.2: 50.4 (#32), Qwen2.5 72B Instruct: 21.4 (#197)
| Benchmark | Muse Spark 1.2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1490 | 1271 |
| DTBench | 94.7% | 62.9% |
| LMCA | 48.4% | 13.4% |
| Epoch Capabilities Index | 155.18 | 129.01 |
| SimpleBench | 74.5% | — |
| NYT Connections (extended) | 79.2% | — |
| CritPt | 17.7% | — |
| BIG-Bench Hard | — | 79.8% |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Muse Spark 1.2 leads
Muse Spark 1.2: 43.8 (#68), Qwen2.5 72B Instruct: 18.1 (#281)
| Benchmark | Muse Spark 1.2 | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| ProofBench | 43% | — |
| Omni-MATH | — | 33% |
| LMArena Math | — | 1283 |
| MATH Level 5 | — | 63.2% |
Knowledge Muse Spark 1.2 leads
Muse Spark 1.2: 53.9 (#49), Qwen2.5 72B Instruct: 26.9 (#247)
| Benchmark | Muse Spark 1.2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Expert | 1485 | 1245 |
| GPQA Diamond | — | 49.1% |
| SimpleQA Verified | 60.3% | — |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multimodal Not comparable
Muse Spark 1.2: 42.6 (#24), Qwen2.5 72B Instruct: —
| Benchmark | Muse Spark 1.2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1304 | — |
Multilingual Muse Spark 1.2 leads
Muse Spark 1.2: 57.4 (#10), Qwen2.5 72B Instruct: 41.2 (#208)
| Benchmark | Muse Spark 1.2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1479 | 1252 |
| LMArena Chinese | 1519 | 1272 |
| LMArena Russian | 1494 | 1264 |
| LMArena French | — | 1279 |
| LMArena German | — | 1234 |
| LMArena Japanese | — | 1180 |
| LMArena Korean | — | 1188 |
| LMArena Spanish | — | 1255 |
Instruction Following Muse Spark 1.2 leads
Muse Spark 1.2: 77.1 (#30), Qwen2.5 72B Instruct: 65.7 (#216)
| Benchmark | Muse Spark 1.2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1466 | 1254 |
| IFEval | — | 80.6% |
Long Context Muse Spark 1.2 leads
Muse Spark 1.2: 45.4 (#35), Qwen2.5 72B Instruct: 38.9 (#183)
| Benchmark | Muse Spark 1.2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1483 | 1282 |
Writing & Preference Muse Spark 1.2 leads
Muse Spark 1.2: 72.6 (#14), Qwen2.5 72B Instruct: 46.9 (#210)
| Benchmark | Muse Spark 1.2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1486 | 1269 |
| LMArena Creative Writing | 1444 | 1221 |
| LMArena Multi-Turn | 1504 | 1272 |
| EQ-Bench Creative Writing | 1840 | — |
| WildBench | — | 80.2% |
Frequently asked questions
Is Muse Spark 1.2 better than Qwen2.5 72B Instruct?
Muse Spark 1.2 is the stronger model overall, scoring 49.8 to 31.5 on the Noometry Index.
Which is cheaper, Muse Spark 1.2 or Qwen2.5 72B Instruct?
Muse Spark 1.2 is cheaper. It lists at $1.25 per million input tokens and $4.25 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Is Muse Spark 1.2 or Qwen2.5 72B Instruct better for coding?
Muse Spark 1.2 scores higher on coding benchmarks: 48.6 versus 32.8 in the Noometry coding category.
Which has the bigger context window?
Muse Spark 1.2 does, with 1.05M tokens against 131K.
How many benchmarks do Muse Spark 1.2 and Qwen2.5 72B Instruct share?
15 benchmarks have published results for both models. Muse Spark 1.2 has 28 scored results on Noometry and Qwen2.5 72B Instruct has 43.