Model comparison
Muse Spark 1.3 vs Qwen3.8 27B
Muse Spark 1.3 is the stronger model overall, scoring 54.8 to 46.0 on the Noometry Index. Qwen3.8 27B costs 3.3× less per token, which makes it the better buy when Muse Spark 1.3's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 28 benchmarks with published results for both. Muse Spark 1.3 scores higher in 10 categories and Qwen3.8 27B in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Muse Spark 1.3 leads 73.1 to 37.1.
- The biggest single-benchmark swing is ProofBench: 58% for Muse Spark 1.3 and 16% for Qwen3.8 27B.
- Qwen3.8 27B is cheaper at $0.04 / $2.30 per million input/output tokens, against $1.25 / $4.25 for Muse Spark 1.3.
- Muse Spark 1.3 accepts more context: 1.05M tokens versus 262K.
- Qwen3.8 27B has downloadable open weights; the other is API-only.
Side by side
| Muse Spark 1.3 | Qwen3.8 27B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 54.8 | 46.0 |
| Released | 2026-09-02 | 2026-08-14 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 131K | 33K |
| Input $ / M tokens | $1.25 | $0.04 |
| Output $ / M tokens | $4.25 | $2.30 |
| Results tracked | 37 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Muse Spark 1.3 leads
Muse Spark 1.3: 56.6 (#21), Qwen3.8 27B: 50.5 (#44)
| Benchmark | Muse Spark 1.3 | Qwen3.8 27B |
|---|---|---|
| LMArena WebDev | 1657 | 1593 |
| SciCode | 59.7% | 46.6% |
| LMArena Coding | 1514 | 1482 |
| CursorBench | 41.6% | — |
Agentic & Tool Use Muse Spark 1.3 leads
Muse Spark 1.3: 38.6 (#30), Qwen3.8 27B: 32.9 (#57)
| Benchmark | Muse Spark 1.3 | Qwen3.8 27B |
|---|---|---|
| APEX-Agents | 57.8% | 47.5% |
| GDP.pdf | 27.6% | — |
Reasoning Muse Spark 1.3 leads
Muse Spark 1.3: 54.0 (#27), Qwen3.8 27B: 41.0 (#54)
| Benchmark | Muse Spark 1.3 | Qwen3.8 27B |
|---|---|---|
| NYT Connections (extended) | 85.1% | 54.5% |
| CritPt | 26% | 5.4% |
| LMArena Hard Prompts | 1503 | 1460 |
| DTBench | 96.5% | 88% |
| LMCA | 53.9% | 41.4% |
| Epoch Capabilities Index | 156.75 | 149.38 |
| ARC-AGI-2 | — | 42.4% |
| ARC-AGI-1 | — | 87.5% |
| Chess Puzzles | 38% | — |
| Mystery Game Puzzles | 25% | — |
| Surface Evolver Bench | — | 45% |
| Bench to the Future 3 | 0.14 | — |
Math Muse Spark 1.3 leads
Muse Spark 1.3: 73.1 (#21), Qwen3.8 27B: 37.1 (#161)
| Benchmark | Muse Spark 1.3 | Qwen3.8 27B |
|---|---|---|
| ProofBench | 58% | 16% |
| LMArena Math | 1494 | 1456 |
| FrontierMath (Tiers 1-3) | 74.4% | — |
| FrontierMath Tier 4 | 46.3% | — |
| OTIS Mock AIME 2024-2025 | 99.2% | — |
Knowledge Muse Spark 1.3 leads
Muse Spark 1.3: 42.6 (#95), Qwen3.8 27B: 41.6 (#109)
| Benchmark | Muse Spark 1.3 | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | 1516 | 1482 |
Multimodal Muse Spark 1.3 leads
Muse Spark 1.3: 43.7 (#22), Qwen3.8 27B: 41.3 (#37)
| Benchmark | Muse Spark 1.3 | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | 1309 | 1271 |
| LMArena Document | 1471 | — |
Multilingual Muse Spark 1.3 leads
Muse Spark 1.3: 57.4 (#8), Qwen3.8 27B: 53.7 (#60)
| Benchmark | Muse Spark 1.3 | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1481 | 1430 |
| LMArena Chinese | 1529 | 1504 |
| LMArena French | 1524 | 1465 |
| LMArena German | 1515 | 1438 |
| LMArena Japanese | 1474 | 1384 |
| LMArena Korean | 1501 | 1393 |
| LMArena Russian | 1490 | 1415 |
| LMArena Spanish | 1490 | 1448 |
Instruction Following Muse Spark 1.3 leads
Muse Spark 1.3: 77.5 (#22), Qwen3.8 27B: 75.8 (#53)
| Benchmark | Muse Spark 1.3 | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1477 | 1439 |
Long Context Muse Spark 1.3 leads
Muse Spark 1.3: 45.6 (#32), Qwen3.8 27B: 44.3 (#70)
| Benchmark | Muse Spark 1.3 | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1488 | 1450 |
Writing & Preference Muse Spark 1.3 leads
Muse Spark 1.3: 73.6 (#9), Qwen3.8 27B: 65.8 (#43)
| Benchmark | Muse Spark 1.3 | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1490 | 1441 |
| LMArena Creative Writing | 1455 | 1384 |
| EQ-Bench Creative Writing | 1906 | 1671 |
| LMArena Multi-Turn | 1482 | 1441 |
Frequently asked questions
Is Muse Spark 1.3 better than Qwen3.8 27B?
Muse Spark 1.3 is the stronger model overall, scoring 54.8 to 46.0 on the Noometry Index. Qwen3.8 27B costs 3.3× less per token, which makes it the better buy when Muse Spark 1.3's lead doesn't matter for your workload.
Which is cheaper, Muse Spark 1.3 or Qwen3.8 27B?
Qwen3.8 27B is cheaper. It lists at $0.04 per million input tokens and $2.30 per million output tokens; Muse Spark 1.3 lists at $1.25 and $4.25.
Is Muse Spark 1.3 or Qwen3.8 27B better for coding?
Muse Spark 1.3 scores higher on coding benchmarks: 56.6 versus 50.5 in the Noometry coding category.
Which has the bigger context window?
Muse Spark 1.3 does, with 1.05M tokens against 262K.
How many benchmarks do Muse Spark 1.3 and Qwen3.8 27B share?
28 benchmarks have published results for both models. Muse Spark 1.3 has 37 scored results on Noometry and Qwen3.8 27B has 31.