Model comparison
Muse Spark 1.2 vs Nemotron 3 Ultra
Muse Spark 1.2 is the stronger model overall, scoring 50.3 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 2.2× 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
NVIDIASummary
- They share 23 benchmarks with published results for both. Muse Spark 1.2 scores higher in 9 categories and Nemotron 3 Ultra in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Muse Spark 1.2 leads 51.3 to 31.0.
- The biggest single-benchmark swing is ProofBench: 43% for Muse Spark 1.2 and 2% for Nemotron 3 Ultra.
- Nemotron 3 Ultra is cheaper at $0.50 / $2.20 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 262K.
- Nemotron 3 Ultra has downloadable open weights; the other is API-only.
Side by side
| Muse Spark 1.2 | Nemotron 3 Ultra | |
|---|---|---|
| Provider | Meta | NVIDIA |
| Noometry Index | 50.3 | 42.5 |
| Released | 2026-08-05 | 2026-06-04 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 131K | 128K |
| Input $ / M tokens | $1.25 | $0.50 |
| Output $ / M tokens | $4.25 | $2.20 |
| Results tracked | 31 | 30 |
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), Nemotron 3 Ultra: 38.1 (#182)
| Benchmark | Muse Spark 1.2 | Nemotron 3 Ultra |
|---|---|---|
| SciCode | 56.4% | 40.3% |
| WeirdML | 60.3% | 43.5% |
| LMArena Coding | 1495 | 1468 |
| DeepSWE | 54.9% | — |
| FrontierCode | — | 13.6% |
| LMArena WebDev | 1533 | — |
| FrontierSWE | 12% | — |
Agentic & Tool Use Muse Spark 1.2 leads
Muse Spark 1.2: 29.4 (#87), Nemotron 3 Ultra: 23.4 (#126)
| Benchmark | Muse Spark 1.2 | Nemotron 3 Ultra |
|---|---|---|
| APEX-Agents | 36.4% | 22.7% |
| GDP.pdf | 16% | — |
Reasoning Muse Spark 1.2 leads
Muse Spark 1.2: 51.3 (#34), Nemotron 3 Ultra: 31.0 (#82)
| Benchmark | Muse Spark 1.2 | Nemotron 3 Ultra |
|---|---|---|
| CritPt | 17.7% | 3.1% |
| LMArena Hard Prompts | 1486 | 1452 |
| DTBench | 94.7% | 90.1% |
| LMCA | 48.4% | 36.9% |
| Epoch Capabilities Index | 154.87 | 146.17 |
| SimpleBench | 74.5% | — |
| NYT Connections (extended) | 79.2% | — |
| Chess Puzzles | — | 12% |
| Mystery Game Puzzles | — | 20% |
Math Muse Spark 1.2 leads
Muse Spark 1.2: 46.4 (#70), Nemotron 3 Ultra: 35.3 (#187)
| Benchmark | Muse Spark 1.2 | Nemotron 3 Ultra |
|---|---|---|
| ProofBench | 43% | 2% |
| LMArena Math | 1471 | 1457 |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
Knowledge Muse Spark 1.2 leads
Muse Spark 1.2: 54.1 (#53), Nemotron 3 Ultra: 52.5 (#61)
| Benchmark | Muse Spark 1.2 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Expert | 1480 | 1472 |
| GPQA Diamond | — | 85.4% |
| SimpleQA Verified | 60.3% | — |
Multimodal Not comparable
Muse Spark 1.2: 43.4 (#25), Nemotron 3 Ultra: —
| Benchmark | Muse Spark 1.2 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Vision | 1305 | — |
Multilingual Muse Spark 1.2 leads
Muse Spark 1.2: 57.1 (#11), Nemotron 3 Ultra: 53.5 (#64)
| Benchmark | Muse Spark 1.2 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1478 | 1427 |
| LMArena Chinese | 1511 | 1497 |
| LMArena French | 1513 | 1472 |
| LMArena Russian | 1487 | 1417 |
| LMArena Spanish | 1498 | 1454 |
| LMArena German | — | 1471 |
| LMArena Korean | — | 1386 |
Instruction Following Muse Spark 1.2 leads
Muse Spark 1.2: 76.7 (#36), Nemotron 3 Ultra: 74.7 (#88)
| Benchmark | Muse Spark 1.2 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1461 | 1418 |
Long Context Muse Spark 1.2 leads
Muse Spark 1.2: 45.2 (#48), Nemotron 3 Ultra: 43.9 (#81)
| Benchmark | Muse Spark 1.2 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1475 | 1437 |
Writing & Preference Muse Spark 1.2 leads
Muse Spark 1.2: 72.3 (#14), Nemotron 3 Ultra: 66.3 (#36)
| Benchmark | Muse Spark 1.2 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1482 | 1445 |
| LMArena Creative Writing | 1449 | 1398 |
| EQ-Bench Creative Writing | 1840 | 1692 |
| LMArena Multi-Turn | 1494 | 1411 |
Frequently asked questions
Is Muse Spark 1.2 better than Nemotron 3 Ultra?
Muse Spark 1.2 is the stronger model overall, scoring 50.3 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 2.2× 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 Nemotron 3 Ultra?
Nemotron 3 Ultra is cheaper. It lists at $0.50 per million input tokens and $2.20 per million output tokens; Muse Spark 1.2 lists at $1.25 and $4.25.
Is Muse Spark 1.2 or Nemotron 3 Ultra better for coding?
Muse Spark 1.2 scores higher on coding benchmarks: 49.2 versus 38.1 in the Noometry coding category.
Which has the bigger context window?
Muse Spark 1.2 does, with 1.05M tokens against 262K.
How many benchmarks do Muse Spark 1.2 and Nemotron 3 Ultra share?
23 benchmarks have published results for both models. Muse Spark 1.2 has 31 scored results on Noometry and Nemotron 3 Ultra has 30.