Model comparison
MiniMax-M3 vs Muse Spark 1.1
Muse Spark 1.1 is the stronger model overall, scoring 49.9 to 43.8 on the Noometry Index. MiniMax-M3 costs 3.8× less per token, which makes it the better buy when Muse Spark 1.1's lead doesn't matter for your workload.
Last verified . 32 shared benchmarks.
MiniMax
MetaSummary
- They share 32 benchmarks with published results for both. MiniMax-M3 scores higher in 1 category and Muse Spark 1.1 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Muse Spark 1.1 leads 47.1 to 30.1.
- The biggest single-benchmark swing is ProofBench: 18% for MiniMax-M3 and 39% for Muse Spark 1.1.
- MiniMax-M3 is cheaper at $0.30 / $1.20 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 1M.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| MiniMax-M3 | Muse Spark 1.1 | |
|---|---|---|
| Provider | MiniMax | Meta |
| Noometry Index | 43.8 | 49.9 |
| Released | 2026-06-01 | 2026-04-08 |
| Weights | Open | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 512K | 131K |
| Input $ / M tokens | $0.30 | $1.25 |
| Output $ / M tokens | $1.20 | $4.25 |
| Results tracked | 41 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Muse Spark 1.1 leads
MiniMax-M3: 41.8 (#118), Muse Spark 1.1: 51.3 (#40)
| Benchmark | MiniMax-M3 | Muse Spark 1.1 |
|---|---|---|
| LMArena WebDev | 1482 | 1542 |
| SciCode | 47.1% | 58.8% |
| LMArena Coding | 1469 | 1498 |
| DeepSWE | — | 53.3% |
| FrontierCode | 14.7% | — |
| ALE-Bench | 640.02 | — |
Agentic & Tool Use Muse Spark 1.1 leads
MiniMax-M3: 22.6 (#130), Muse Spark 1.1: 30.8 (#73)
| Benchmark | MiniMax-M3 | Muse Spark 1.1 |
|---|---|---|
| APEX-Agents | 37.7% | 31.8% |
| GBAEval | 0.9% | 7.9% |
| Vending-Bench 2 | 2,158 | 6,520 |
| OSWorld 2.0 | 4.6% | — |
| τ²-bench Banking | — | 40.5% |
| GDP.pdf | — | 15% |
Reasoning Muse Spark 1.1 leads
MiniMax-M3: 30.1 (#87), Muse Spark 1.1: 47.1 (#44)
| Benchmark | MiniMax-M3 | Muse Spark 1.1 |
|---|---|---|
| NYT Connections (extended) | 65.1% | 84.9% |
| CritPt | 3.7% | 15.1% |
| LMArena Hard Prompts | 1447 | 1486 |
| DTBench | 78.9% | 94.4% |
| LMCA | 33.7% | 49.9% |
| Surface Evolver Bench | 55% | 52.5% |
| Epoch Capabilities Index | 146.95 | 154.21 |
| SimpleBench | 45.8% | — |
| Chess Puzzles | 14% | — |
| Mystery Game Puzzles | 8% | — |
| ForecastBench | 61.4 | — |
Math Muse Spark 1.1 leads
MiniMax-M3: 40.0 (#95), Muse Spark 1.1: 45.5 (#76)
| Benchmark | MiniMax-M3 | Muse Spark 1.1 |
|---|---|---|
| ProofBench | 18% | 39% |
| LMArena Math | 1429 | 1483 |
| OTIS Mock AIME 2024-2025 | 71.1% | — |
Knowledge MiniMax-M3 leads
MiniMax-M3: 58.4 (#35), Muse Spark 1.1: 53.1 (#59)
| Benchmark | MiniMax-M3 | Muse Spark 1.1 |
|---|---|---|
| LMArena Expert | 1461 | 1478 |
| GPQA Diamond | 90.9% | — |
| SimpleQA Verified | — | 57.8% |
Multimodal Muse Spark 1.1 leads
MiniMax-M3: 40.2 (#51), Muse Spark 1.1: 42.6 (#29)
| Benchmark | MiniMax-M3 | Muse Spark 1.1 |
|---|---|---|
| LMArena Vision | 1253 | 1293 |
| LMArena Document | 1435 | 1465 |
Multilingual Muse Spark 1.1 leads
MiniMax-M3: 53.0 (#75), Muse Spark 1.1: 56.7 (#17)
| Benchmark | MiniMax-M3 | Muse Spark 1.1 |
|---|---|---|
| LMArena Non-English | 1420 | 1472 |
| LMArena Chinese | 1463 | 1518 |
| LMArena French | 1447 | 1494 |
| LMArena German | 1426 | 1466 |
| LMArena Japanese | 1381 | 1451 |
| LMArena Korean | 1372 | 1458 |
| LMArena Russian | 1428 | 1483 |
| LMArena Spanish | 1432 | 1464 |
Instruction Following Muse Spark 1.1 leads
MiniMax-M3: 75.5 (#62), Muse Spark 1.1: 76.5 (#39)
| Benchmark | MiniMax-M3 | Muse Spark 1.1 |
|---|---|---|
| LMArena Instruction Following | 1433 | 1457 |
Long Context Too close to call
MiniMax-M3: 44.2 (#72), Muse Spark 1.1: 44.8 (#58)
| Benchmark | MiniMax-M3 | Muse Spark 1.1 |
|---|---|---|
| LMArena Longer Query | 1445 | 1462 |
Writing & Preference Muse Spark 1.1 leads
MiniMax-M3: 62.1 (#83), Muse Spark 1.1: 73.4 (#11)
| Benchmark | MiniMax-M3 | Muse Spark 1.1 |
|---|---|---|
| LMArena Text | 1433 | 1479 |
| LMArena Creative Writing | 1404 | 1437 |
| EQ-Bench 4 | 1150 | 1260 |
| LMArena Multi-Turn | 1442 | 1485 |
| EQ-Bench Creative Writing | — | 1927 |
Frequently asked questions
Is MiniMax-M3 better than Muse Spark 1.1?
Muse Spark 1.1 is the stronger model overall, scoring 49.9 to 43.8 on the Noometry Index. MiniMax-M3 costs 3.8× 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, MiniMax-M3 or Muse Spark 1.1?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Muse Spark 1.1 lists at $1.25 and $4.25.
Is MiniMax-M3 or Muse Spark 1.1 better for coding?
Muse Spark 1.1 scores higher on coding benchmarks: 51.3 versus 41.8 in the Noometry coding category.
Which has the bigger context window?
Muse Spark 1.1 does, with 1.05M tokens against 1M.
How many benchmarks do MiniMax-M3 and Muse Spark 1.1 share?
32 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Muse Spark 1.1 has 37.