Model comparison
MiniMax-M3 vs Solar Pro4
MiniMax-M3 is the stronger model overall, scoring 43.8 to 42.1 on the Noometry Index.
Last verified . 18 shared benchmarks.
MiniMax
UpstageSummary
- They share 18 benchmarks with published results for both. MiniMax-M3 scores higher in 8 categories and Solar Pro4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 39.8.
- Both cost about the same: $0.30 input and $1.20 output per million tokens.
- MiniMax-M3 accepts more context: 1M tokens versus 524K.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| MiniMax-M3 | Solar Pro4 | |
|---|---|---|
| Provider | MiniMax | Upstage |
| Noometry Index | 43.8 | 42.1 |
| Released | 2026-06-01 | 2026-08-06 |
| Weights | Open | Proprietary |
| Context window | 1M | 524K |
| Max output | 512K | 131K |
| Input $ / M tokens | $0.30 | $0.30 |
| Output $ / M tokens | $1.20 | $1.20 |
| Results tracked | 41 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
MiniMax-M3: 41.8 (#118), Solar Pro4: 40.1 (#149)
| Benchmark | MiniMax-M3 | Solar Pro4 |
|---|---|---|
| LMArena WebDev | 1482 | 1371 |
| LMArena Coding | 1469 | 1437 |
| FrontierCode | 14.7% | — |
| SciCode | 47.1% | — |
| ALE-Bench | 640.02 | — |
Agentic & Tool Use Not comparable
MiniMax-M3: 22.6 (#130), Solar Pro4: —
| Benchmark | MiniMax-M3 | Solar Pro4 |
|---|---|---|
| APEX-Agents | 37.7% | — |
| OSWorld 2.0 | 4.6% | — |
| GBAEval | 0.9% | — |
| Vending-Bench 2 | 2,158 | — |
Reasoning MiniMax-M3 leads
MiniMax-M3: 30.1 (#87), Solar Pro4: 28.5 (#104)
| Benchmark | MiniMax-M3 | Solar Pro4 |
|---|---|---|
| LMArena Hard Prompts | 1447 | 1399 |
| SimpleBench | 45.8% | — |
| NYT Connections (extended) | 65.1% | — |
| CritPt | 3.7% | — |
| Chess Puzzles | 14% | — |
| Mystery Game Puzzles | 8% | — |
| DTBench | 78.9% | — |
| LMCA | 33.7% | — |
| Surface Evolver Bench | 55% | — |
| Epoch Capabilities Index | 146.95 | — |
| ForecastBench | 61.4 | — |
Math MiniMax-M3 leads
MiniMax-M3: 40.0 (#95), Solar Pro4: 38.8 (#128)
| Benchmark | MiniMax-M3 | Solar Pro4 |
|---|---|---|
| LMArena Math | 1429 | 1416 |
| OTIS Mock AIME 2024-2025 | 71.1% | — |
| ProofBench | 18% | — |
Knowledge MiniMax-M3 leads
MiniMax-M3: 58.4 (#35), Solar Pro4: 39.8 (#129)
| Benchmark | MiniMax-M3 | Solar Pro4 |
|---|---|---|
| LMArena Expert | 1461 | 1427 |
| GPQA Diamond | 90.9% | — |
Multimodal Not comparable
MiniMax-M3: 40.2 (#51), Solar Pro4: —
| Benchmark | MiniMax-M3 | Solar Pro4 |
|---|---|---|
| LMArena Vision | 1253 | — |
| LMArena Document | 1435 | — |
Multilingual MiniMax-M3 leads
MiniMax-M3: 53.0 (#75), Solar Pro4: 48.7 (#139)
| Benchmark | MiniMax-M3 | Solar Pro4 |
|---|---|---|
| LMArena Non-English | 1420 | 1361 |
| LMArena Chinese | 1463 | 1415 |
| LMArena French | 1447 | 1397 |
| LMArena German | 1426 | 1364 |
| LMArena Japanese | 1381 | 1309 |
| LMArena Korean | 1372 | 1382 |
| LMArena Russian | 1428 | 1360 |
| LMArena Spanish | 1432 | 1401 |
Instruction Following MiniMax-M3 leads
MiniMax-M3: 75.5 (#62), Solar Pro4: 72.7 (#132)
| Benchmark | MiniMax-M3 | Solar Pro4 |
|---|---|---|
| LMArena Instruction Following | 1433 | 1377 |
Long Context MiniMax-M3 leads
MiniMax-M3: 44.2 (#72), Solar Pro4: 42.1 (#130)
| Benchmark | MiniMax-M3 | Solar Pro4 |
|---|---|---|
| LMArena Longer Query | 1445 | 1381 |
Writing & Preference MiniMax-M3 leads
MiniMax-M3: 62.1 (#83), Solar Pro4: 56.5 (#138)
| Benchmark | MiniMax-M3 | Solar Pro4 |
|---|---|---|
| LMArena Text | 1433 | 1386 |
| LMArena Creative Writing | 1404 | 1316 |
| LMArena Multi-Turn | 1442 | 1385 |
| EQ-Bench 4 | 1150 | — |
Frequently asked questions
Is MiniMax-M3 better than Solar Pro4?
MiniMax-M3 is the stronger model overall, scoring 43.8 to 42.1 on the Noometry Index.
Which is cheaper, MiniMax-M3 or Solar Pro4?
Solar Pro4 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.
Is MiniMax-M3 or Solar Pro4 better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 40.1 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 524K.
How many benchmarks do MiniMax-M3 and Solar Pro4 share?
18 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Solar Pro4 has 18.