Model comparison
MiniMax-M2 vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 37.4 on the Noometry Index.
Last verified . 15 shared benchmarks.
MiniMax
StepFunSummary
- They share 15 benchmarks with published results for both. MiniMax-M2 scores higher in 2 categories and Step 3 in 6 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 3 leads 28.4 to 19.4.
Side by side
| MiniMax-M2 | Step 3 | |
|---|---|---|
| Provider | MiniMax | StepFun |
| Noometry Index | 37.4 | 40.5 |
| Released | 2025-10-27 | — |
| Weights | Open | Open |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 21 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
MiniMax-M2: 39.3 (#159), Step 3: 40.1 (#147)
| Benchmark | MiniMax-M2 | Step 3 |
|---|---|---|
| LMArena Coding | 1370 | 1367 |
| SWE-bench Verified (bash only) | 61% | — |
| LMArena WebDev | 1297 | — |
Agentic & Tool Use Not comparable
MiniMax-M2: 25.1 (#109), Step 3: —
| Benchmark | MiniMax-M2 | Step 3 |
|---|---|---|
| Terminal-Bench | 30% | — |
| Vending-Bench 2 | 160.6 | — |
Reasoning Step 3 leads
MiniMax-M2: 19.4 (#258), Step 3: 28.4 (#105)
| Benchmark | MiniMax-M2 | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 57.8% | 62.3% |
| LMArena Hard Prompts | 1357 | 1355 |
| NYT Connections (extended) | 14.8% | — |
Math Too close to call
MiniMax-M2: 37.3 (#160), Step 3: 37.6 (#148)
| Benchmark | MiniMax-M2 | Step 3 |
|---|---|---|
| LMArena Math | 1352 | 1366 |
Knowledge Too close to call
MiniMax-M2: 37.0 (#163), Step 3: 36.8 (#164)
| Benchmark | MiniMax-M2 | Step 3 |
|---|---|---|
| LMArena Expert | 1337 | 1333 |
Multimodal Not comparable
MiniMax-M2: —, Step 3: 35.5 (#86)
| Benchmark | MiniMax-M2 | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual Step 3 leads
MiniMax-M2: 45.3 (#171), Step 3: 46.3 (#159)
| Benchmark | MiniMax-M2 | Step 3 |
|---|---|---|
| LMArena Non-English | 1313 | 1327 |
| LMArena Chinese | 1366 | 1397 |
| LMArena German | 1355 | 1371 |
| LMArena Russian | 1331 | 1331 |
| LMArena Spanish | 1326 | 1371 |
| LMArena French | 1335 | — |
| LMArena Korean | — | 1269 |
Instruction Following Too close to call
MiniMax-M2: 70.2 (#166), Step 3: 70.4 (#164)
| Benchmark | MiniMax-M2 | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1328 | 1332 |
Long Context Too close to call
MiniMax-M2: 40.5 (#153), Step 3: 40.3 (#157)
| Benchmark | MiniMax-M2 | Step 3 |
|---|---|---|
| LMArena Longer Query | 1331 | 1326 |
Writing & Preference Step 3 leads
MiniMax-M2: 53.0 (#162), Step 3: 54.3 (#151)
| Benchmark | MiniMax-M2 | Step 3 |
|---|---|---|
| LMArena Text | 1340 | 1350 |
| LMArena Creative Writing | 1286 | 1321 |
| LMArena Multi-Turn | 1361 | 1341 |
Frequently asked questions
Is MiniMax-M2 better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 37.4 on the Noometry Index.
Is MiniMax-M2 or Step 3 better for coding?
They score almost the same on coding (39.3 vs 40.1); test both on your own repository before choosing.
How many benchmarks do MiniMax-M2 and Step 3 share?
15 benchmarks have published results for both models. MiniMax-M2 has 21 scored results on Noometry and Step 3 has 17.