Model comparison
MiniMax-M2.1 vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 38.9 on the Noometry Index.
Last verified . 15 shared benchmarks.
MiniMax
StepFunSummary
- They share 15 benchmarks with published results for both. MiniMax-M2.1 scores higher in 7 categories and Step 3 in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 3 leads 28.4 to 16.6.
Side by side
| MiniMax-M2.1 | Step 3 | |
|---|---|---|
| Provider | MiniMax | StepFun |
| Noometry Index | 38.9 | 40.5 |
| Released | 2025-12-23 | — |
| Weights | Open | Open |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 22 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
MiniMax-M2.1: 40.4 (#143), Step 3: 40.1 (#147)
| Benchmark | MiniMax-M2.1 | Step 3 |
|---|---|---|
| LMArena Coding | 1421 | 1367 |
| LMArena WebDev | 1384 | — |
| ALE-Bench | 623.83 | — |
Agentic & Tool Use Not comparable
MiniMax-M2.1: 27.9 (#98), Step 3: —
| Benchmark | MiniMax-M2.1 | Step 3 |
|---|---|---|
| Terminal-Bench | 36.6% | — |
Reasoning Step 3 leads
MiniMax-M2.1: 16.6 (#302), Step 3: 28.4 (#105)
| Benchmark | MiniMax-M2.1 | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1411 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
| NYT Connections (extended) | 11.2% | — |
Math Too close to call
MiniMax-M2.1: 38.3 (#138), Step 3: 37.6 (#148)
| Benchmark | MiniMax-M2.1 | Step 3 |
|---|---|---|
| LMArena Math | 1397 | 1366 |
Knowledge MiniMax-M2.1 leads
MiniMax-M2.1: 38.3 (#147), Step 3: 36.8 (#164)
| Benchmark | MiniMax-M2.1 | Step 3 |
|---|---|---|
| LMArena Expert | 1431 | 1333 |
| Vectara Hallucination Rate | 11.8% | — |
Multimodal Not comparable
MiniMax-M2.1: —, Step 3: 35.5 (#86)
| Benchmark | MiniMax-M2.1 | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual MiniMax-M2.1 leads
MiniMax-M2.1: 50.0 (#128), Step 3: 46.3 (#159)
| Benchmark | MiniMax-M2.1 | Step 3 |
|---|---|---|
| LMArena Non-English | 1378 | 1327 |
| LMArena Chinese | 1430 | 1397 |
| LMArena German | 1381 | 1371 |
| LMArena Korean | 1298 | 1269 |
| LMArena Russian | 1387 | 1331 |
| LMArena Spanish | 1397 | 1371 |
| LMArena French | 1404 | — |
| LMArena Japanese | 1287 | — |
Instruction Following MiniMax-M2.1 leads
MiniMax-M2.1: 73.8 (#112), Step 3: 70.4 (#164)
| Benchmark | MiniMax-M2.1 | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1400 | 1332 |
Long Context MiniMax-M2.1 leads
MiniMax-M2.1: 43.2 (#101), Step 3: 40.3 (#157)
| Benchmark | MiniMax-M2.1 | Step 3 |
|---|---|---|
| LMArena Longer Query | 1416 | 1326 |
Writing & Preference MiniMax-M2.1 leads
MiniMax-M2.1: 58.3 (#120), Step 3: 54.3 (#151)
| Benchmark | MiniMax-M2.1 | Step 3 |
|---|---|---|
| LMArena Text | 1392 | 1350 |
| LMArena Creative Writing | 1361 | 1321 |
| LMArena Multi-Turn | 1396 | 1341 |
Frequently asked questions
Is MiniMax-M2.1 better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 38.9 on the Noometry Index.
Is MiniMax-M2.1 or Step 3 better for coding?
They score almost the same on coding (40.4 vs 40.1); test both on your own repository before choosing.
How many benchmarks do MiniMax-M2.1 and Step 3 share?
15 benchmarks have published results for both models. MiniMax-M2.1 has 22 scored results on Noometry and Step 3 has 17.