Model comparison
MiniMax-M2.5 vs Muse Spark
Muse Spark is the stronger model overall, scoring 50.6 to 38.3 on the Noometry Index.
Last verified . 18 shared benchmarks.
MiniMax
MetaSummary
- They share 18 benchmarks with published results for both. MiniMax-M2.5 scores higher in 1 category and Muse Spark in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Muse Spark leads 65.7 to 39.2.
- The biggest single-benchmark swing is ProofBench: 4% for MiniMax-M2.5 and 17% for Muse Spark.
- MiniMax-M2.5 has downloadable open weights; the other is API-only.
Side by side
| MiniMax-M2.5 | Muse Spark | |
|---|---|---|
| Provider | MiniMax | Meta |
| Noometry Index | 38.3 | 50.6 |
| Released | 2026-02-12 | 2026-04-08 |
| Weights | Open | Proprietary |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 33 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
MiniMax-M2.5: 48.1 (#58), Muse Spark: 46.2 (#69)
| Benchmark | MiniMax-M2.5 | Muse Spark |
|---|---|---|
| LMArena Coding | 1381 | 1481 |
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1387 | — |
| SWE-bench Multilingual | 68.3% | — |
| SciCode | — | 51.5% |
| ALE-Bench | 618.17 | — |
Agentic & Tool Use Not comparable
MiniMax-M2.5: 30.4 (#77), Muse Spark: —
| Benchmark | MiniMax-M2.5 | Muse Spark |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Vending-Bench 2 | -23.16 | — |
Reasoning Muse Spark leads
MiniMax-M2.5: 17.5 (#292), Muse Spark: 35.9 (#67)
| Benchmark | MiniMax-M2.5 | Muse Spark |
|---|---|---|
| LMArena Hard Prompts | 1372 | 1474 |
| Epoch Capabilities Index | 146.68 | 152.04 |
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 55.2% | — |
| NYT Connections (extended) | 16.8% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | — | 11.3% |
Math Muse Spark leads
MiniMax-M2.5: 26.9 (#253), Muse Spark: 47.8 (#66)
| Benchmark | MiniMax-M2.5 | Muse Spark |
|---|---|---|
| ProofBench | 4% | 17% |
| LMArena Math | 1378 | 1455 |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Muse Spark leads
MiniMax-M2.5: 39.2 (#135), Muse Spark: 65.7 (#13)
| Benchmark | MiniMax-M2.5 | Muse Spark |
|---|---|---|
| LMArena Expert | 1379 | 1457 |
| GPQA Diamond | — | 89.8% |
| Humanity's Last Exam | — | 40.6% |
| Vectara Hallucination Rate | 9.1% | — |
Multimodal Not comparable
MiniMax-M2.5: —, Muse Spark: 43.4 (#24)
| Benchmark | MiniMax-M2.5 | Muse Spark |
|---|---|---|
| LMArena Vision | — | 1306 |
| LMArena Document | — | 1444 |
Multilingual Muse Spark leads
MiniMax-M2.5: 47.1 (#152), Muse Spark: 56.1 (#24)
| Benchmark | MiniMax-M2.5 | Muse Spark |
|---|---|---|
| LMArena Non-English | 1338 | 1464 |
| LMArena Chinese | 1393 | 1509 |
| LMArena French | 1362 | 1497 |
| LMArena German | 1362 | 1497 |
| LMArena Korean | 1232 | 1459 |
| LMArena Russian | 1358 | 1466 |
| LMArena Spanish | 1354 | 1472 |
| LMArena Japanese | 1171 | — |
Instruction Following Muse Spark leads
MiniMax-M2.5: 71.5 (#148), Muse Spark: 75.9 (#51)
| Benchmark | MiniMax-M2.5 | Muse Spark |
|---|---|---|
| LMArena Instruction Following | 1353 | 1442 |
Long Context Muse Spark leads
MiniMax-M2.5: 37.5 (#216), Muse Spark: 44.4 (#69)
| Benchmark | MiniMax-M2.5 | Muse Spark |
|---|---|---|
| LMArena Longer Query | 1366 | 1451 |
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |
Writing & Preference Muse Spark leads
MiniMax-M2.5: 53.9 (#153), Muse Spark: 66.0 (#39)
| Benchmark | MiniMax-M2.5 | Muse Spark |
|---|---|---|
| LMArena Text | 1359 | 1474 |
| LMArena Creative Writing | 1331 | 1459 |
| LMArena Multi-Turn | 1364 | 1477 |
| EQ-Bench Creative Writing | 1361 | — |
Frequently asked questions
Is MiniMax-M2.5 better than Muse Spark?
Muse Spark is the stronger model overall, scoring 50.6 to 38.3 on the Noometry Index.
Is MiniMax-M2.5 or Muse Spark better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 46.2 in the Noometry coding category.
How many benchmarks do MiniMax-M2.5 and Muse Spark share?
18 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Muse Spark has 27.