Model comparison
Gemma 3 4B vs Muse Spark 1.1
Muse Spark 1.1 is the stronger model overall, scoring 49.9 to 28.1 on the Noometry Index. Gemma 3 4B costs 40× less per token, which makes it the better buy when Muse Spark 1.1's lead doesn't matter for your workload.
Last verified . 16 shared benchmarks.
Google
MetaSummary
- They share 16 benchmarks with published results for both. Gemma 3 4B scores higher in 0 categories and Muse Spark 1.1 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Muse Spark 1.1 leads 53.1 to 11.8.
- The biggest single-benchmark swing is LMCA: 2.8% for Gemma 3 4B and 49.9% for Muse Spark 1.1.
- Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $1.25 / $4.25 for Muse Spark 1.1.
- Muse Spark 1.1 accepts more context: 1.05M tokens versus 131K.
- Gemma 3 4B has downloadable open weights; the other is API-only.
Side by side
| Gemma 3 4B | Muse Spark 1.1 | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 28.1 | 49.9 |
| Released | 2025-03-12 | 2026-04-08 |
| Weights | Open | Proprietary |
| Context window | 131K | 1.05M |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.04 | $1.25 |
| Output $ / M tokens | $0.08 | $4.25 |
| Results tracked | 22 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Muse Spark 1.1 leads
Gemma 3 4B: 35.9 (#215), Muse Spark 1.1: 51.3 (#40)
| Benchmark | Gemma 3 4B | Muse Spark 1.1 |
|---|---|---|
| LMArena Coding | 1230 | 1498 |
| DeepSWE | — | 53.3% |
| LMArena WebDev | — | 1542 |
| SciCode | — | 58.8% |
Agentic & Tool Use Muse Spark 1.1 leads
Gemma 3 4B: 20.9 (#142), Muse Spark 1.1: 30.8 (#73)
| Benchmark | Gemma 3 4B | Muse Spark 1.1 |
|---|---|---|
| APEX-Agents | — | 31.8% |
| Berkeley Function Calling Leaderboard | 19.6% | — |
| τ²-bench Banking | — | 40.5% |
| GBAEval | — | 7.9% |
| GDP.pdf | — | 15% |
| Vending-Bench 2 | — | 6,520 |
Reasoning Muse Spark 1.1 leads
Gemma 3 4B: 13.2 (#335), Muse Spark 1.1: 47.1 (#44)
| Benchmark | Gemma 3 4B | Muse Spark 1.1 |
|---|---|---|
| LMArena Hard Prompts | 1253 | 1486 |
| DTBench | 50.9% | 94.4% |
| LMCA | 2.8% | 49.9% |
| Epoch Capabilities Index | 116.02 | 154.21 |
| Kagi LLM Benchmark | 25.2% | — |
| NYT Connections (extended) | — | 84.9% |
| CritPt | — | 15.1% |
| Chess Puzzles | 0% | — |
| Surface Evolver Bench | — | 52.5% |
Math Muse Spark 1.1 leads
Gemma 3 4B: 16.8 (#292), Muse Spark 1.1: 45.5 (#76)
| Benchmark | Gemma 3 4B | Muse Spark 1.1 |
|---|---|---|
| LMArena Math | 1239 | 1483 |
| OTIS Mock AIME 2024-2025 | 7.5% | — |
| ProofBench | — | 39% |
Knowledge Muse Spark 1.1 leads
Gemma 3 4B: 11.8 (#299), Muse Spark 1.1: 53.1 (#59)
| Benchmark | Gemma 3 4B | Muse Spark 1.1 |
|---|---|---|
| LMArena Expert | 1223 | 1478 |
| GPQA Diamond | 23.2% | — |
| SimpleQA Verified | — | 57.8% |
| Vectara Hallucination Rate | 6.4% | — |
Multimodal Not comparable
Gemma 3 4B: —, Muse Spark 1.1: 42.6 (#29)
| Benchmark | Gemma 3 4B | Muse Spark 1.1 |
|---|---|---|
| LMArena Vision | — | 1293 |
| LMArena Document | — | 1465 |
Multilingual Muse Spark 1.1 leads
Gemma 3 4B: 42.5 (#194), Muse Spark 1.1: 56.7 (#17)
| Benchmark | Gemma 3 4B | Muse Spark 1.1 |
|---|---|---|
| LMArena Non-English | 1273 | 1472 |
| LMArena German | 1281 | 1466 |
| LMArena Russian | 1294 | 1483 |
| LMArena Chinese | — | 1518 |
| LMArena French | — | 1494 |
| LMArena Japanese | — | 1451 |
| LMArena Korean | — | 1458 |
| LMArena Spanish | — | 1464 |
Instruction Following Muse Spark 1.1 leads
Gemma 3 4B: 65.2 (#225), Muse Spark 1.1: 76.5 (#39)
| Benchmark | Gemma 3 4B | Muse Spark 1.1 |
|---|---|---|
| LMArena Instruction Following | 1239 | 1457 |
Long Context Muse Spark 1.1 leads
Gemma 3 4B: 38.7 (#194), Muse Spark 1.1: 44.8 (#58)
| Benchmark | Gemma 3 4B | Muse Spark 1.1 |
|---|---|---|
| LMArena Longer Query | 1273 | 1462 |
Writing & Preference Muse Spark 1.1 leads
Gemma 3 4B: 42.0 (#239), Muse Spark 1.1: 73.4 (#11)
| Benchmark | Gemma 3 4B | Muse Spark 1.1 |
|---|---|---|
| LMArena Text | 1291 | 1479 |
| LMArena Creative Writing | 1271 | 1437 |
| EQ-Bench Creative Writing | 1068 | 1927 |
| LMArena Multi-Turn | 1255 | 1485 |
| EQ-Bench 4 | — | 1260 |
Frequently asked questions
Is Gemma 3 4B better than Muse Spark 1.1?
Muse Spark 1.1 is the stronger model overall, scoring 49.9 to 28.1 on the Noometry Index. Gemma 3 4B costs 40× less per token, which makes it the better buy when Muse Spark 1.1's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 4B or Muse Spark 1.1?
Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; Muse Spark 1.1 lists at $1.25 and $4.25.
Is Gemma 3 4B or Muse Spark 1.1 better for coding?
Muse Spark 1.1 scores higher on coding benchmarks: 51.3 versus 35.9 in the Noometry coding category.
Which has the bigger context window?
Muse Spark 1.1 does, with 1.05M tokens against 131K.
How many benchmarks do Gemma 3 4B and Muse Spark 1.1 share?
16 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Muse Spark 1.1 has 37.