Model comparison
Muse Spark 1.2 vs Qwen3 235B-A22B
Muse Spark 1.2 is the stronger model overall, scoring 50.3 to 43.5 on the Noometry Index. Qwen3 235B-A22B costs 1.6× less per token, which makes it the better buy when Muse Spark 1.2's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Muse Spark 1.2 scores higher in 6 categories and Qwen3 235B-A22B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Muse Spark 1.2 leads 51.3 to 15.7.
- The biggest single-benchmark swing is SimpleBench: 74.5% for Muse Spark 1.2 and 31% for Qwen3 235B-A22B.
- Qwen3 235B-A22B is cheaper at $0.70 / $2.80 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 131K.
- Qwen3 235B-A22B has downloadable open weights; the other is API-only.
Side by side
| Muse Spark 1.2 | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 50.3 | 43.5 |
| Released | 2026-08-05 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 131K | 16K |
| Input $ / M tokens | $1.25 | $0.70 |
| Output $ / M tokens | $4.25 | $2.80 |
| Results tracked | 31 | 49 |
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: 49.2 (#51), Qwen3 235B-A22B: 44.3 (#75)
| Benchmark | Muse Spark 1.2 | Qwen3 235B-A22B |
|---|---|---|
| SciCode | 56.4% | 42.4% |
| WeirdML | 60.3% | 41% |
| LMArena Coding | 1495 | 1445 |
| DeepSWE | 54.9% | — |
| Aider Polyglot | — | 59.6% |
| LMArena WebDev | 1533 | — |
| FrontierSWE | 12% | — |
Agentic & Tool Use Qwen3 235B-A22B leads
Muse Spark 1.2: 29.4 (#87), Qwen3 235B-A22B: 33.9 (#51)
| Benchmark | Muse Spark 1.2 | Qwen3 235B-A22B |
|---|---|---|
| APEX-Agents | 36.4% | — |
| Berkeley Function Calling Leaderboard | — | 52.1% |
| GDP.pdf | 16% | — |
| Vending-Bench 2 | — | -11.34 |
Reasoning Muse Spark 1.2 leads
Muse Spark 1.2: 51.3 (#34), Qwen3 235B-A22B: 15.7 (#311)
| Benchmark | Muse Spark 1.2 | Qwen3 235B-A22B |
|---|---|---|
| SimpleBench | 74.5% | 31% |
| CritPt | 17.7% | 0% |
| LMArena Hard Prompts | 1486 | 1433 |
| DTBench | 94.7% | 80.3% |
| LMCA | 48.4% | 29.3% |
| Epoch Capabilities Index | 154.87 | 143.85 |
| ARC-AGI-2 | — | 1.3% |
| Kagi LLM Benchmark | — | 69.4% |
| NYT Connections (extended) | 79.2% | — |
| ARC-AGI-1 | — | 11% |
| Chess Puzzles | — | 12% |
| Mystery Game Puzzles | — | 9% |
| ForecastBench | — | 59.7 |
Math Qwen3 235B-A22B leads
Muse Spark 1.2: 46.4 (#70), Qwen3 235B-A22B: 50.4 (#57)
| Benchmark | Muse Spark 1.2 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Math | 1471 | 1432 |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| ProofBench | 43% | — |
| Omni-MATH | — | 71.8% |
| MATH Level 5 | — | 68.9% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Muse Spark 1.2 leads
Muse Spark 1.2: 54.1 (#53), Qwen3 235B-A22B: 49.6 (#73)
| Benchmark | Muse Spark 1.2 | Qwen3 235B-A22B |
|---|---|---|
| SimpleQA Verified | 60.3% | 40.4% |
| LMArena Expert | 1480 | 1463 |
| GPQA Diamond | — | 80.1% |
| MMLU-Pro | — | 84.4% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 72.7% |
Multimodal Not comparable
Muse Spark 1.2: 43.4 (#25), Qwen3 235B-A22B: —
| Benchmark | Muse Spark 1.2 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Vision | 1305 | — |
Multilingual Muse Spark 1.2 leads
Muse Spark 1.2: 57.1 (#11), Qwen3 235B-A22B: 52.3 (#89)
| Benchmark | Muse Spark 1.2 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1478 | 1409 |
| LMArena Chinese | 1511 | 1481 |
| LMArena French | 1513 | 1445 |
| LMArena Russian | 1487 | 1411 |
| LMArena Spanish | 1498 | 1430 |
| LMArena German | — | 1433 |
| LMArena Japanese | — | 1399 |
| LMArena Korean | — | 1391 |
Instruction Following Muse Spark 1.2 leads
Muse Spark 1.2: 76.7 (#36), Qwen3 235B-A22B: 72.6 (#136)
| Benchmark | Muse Spark 1.2 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1461 | 1408 |
| IFEval | — | 83.5% |
Long Context Too close to call
Muse Spark 1.2: 45.2 (#48), Qwen3 235B-A22B: 46.1 (#26)
| Benchmark | Muse Spark 1.2 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1475 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Muse Spark 1.2 leads
Muse Spark 1.2: 72.3 (#14), Qwen3 235B-A22B: 59.6 (#108)
| Benchmark | Muse Spark 1.2 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1482 | 1419 |
| LMArena Creative Writing | 1449 | 1384 |
| EQ-Bench Creative Writing | 1840 | 1366 |
| LMArena Multi-Turn | 1494 | 1432 |
| Short-Story Creative Writing | — | 83% |
| WildBench | — | 86.6% |
Frequently asked questions
Is Muse Spark 1.2 better than Qwen3 235B-A22B?
Muse Spark 1.2 is the stronger model overall, scoring 50.3 to 43.5 on the Noometry Index. Qwen3 235B-A22B costs 1.6× 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, Muse Spark 1.2 or Qwen3 235B-A22B?
Qwen3 235B-A22B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; Muse Spark 1.2 lists at $1.25 and $4.25.
Is Muse Spark 1.2 or Qwen3 235B-A22B better for coding?
Muse Spark 1.2 scores higher on coding benchmarks: 49.2 versus 44.3 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 Qwen3 235B-A22B share?
23 benchmarks have published results for both models. Muse Spark 1.2 has 31 scored results on Noometry and Qwen3 235B-A22B has 49.