Model comparison
MiniMax-M2 vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 37.4 on the Noometry Index. MiniMax-M2 costs 2.7× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Last verified . 14 shared benchmarks.
MiniMax
StepFunSummary
- They share 14 benchmarks with published results for both. MiniMax-M2 scores higher in 0 categories and Step 5 Preview in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 5 Preview leads 40.0 to 19.4.
- MiniMax-M2 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
- Step 5 Preview accepts more context: 1.02M tokens versus 205K.
- MiniMax-M2 has downloadable open weights; the other is API-only.
Side by side
| MiniMax-M2 | Step 5 Preview | |
|---|---|---|
| Provider | MiniMax | StepFun |
| Noometry Index | 37.4 | 47.9 |
| Released | 2025-10-27 | 2026-09-16 |
| Weights | Open | Proprietary |
| Context window | 205K | 1.02M |
| Max output | 131K | 66K |
| Input $ / M tokens | $0.30 | $1 |
| Output $ / M tokens | $1.20 | $2.70 |
| Results tracked | 21 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
MiniMax-M2: 39.3 (#159), Step 5 Preview: 51.8 (#36)
| Benchmark | MiniMax-M2 | Step 5 Preview |
|---|---|---|
| LMArena WebDev | 1297 | 1564 |
| LMArena Coding | 1370 | 1480 |
| SWE-bench Verified (bash only) | 61% | — |
| SciCode | — | 58.9% |
Agentic & Tool Use Not comparable
MiniMax-M2: 25.1 (#109), Step 5 Preview: —
| Benchmark | MiniMax-M2 | Step 5 Preview |
|---|---|---|
| Terminal-Bench | 30% | — |
| Vending-Bench 2 | 160.6 | — |
Reasoning Step 5 Preview leads
MiniMax-M2: 19.4 (#258), Step 5 Preview: 40.0 (#57)
| Benchmark | MiniMax-M2 | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1357 | 1465 |
| Kagi LLM Benchmark | 57.8% | — |
| NYT Connections (extended) | 14.8% | — |
| CritPt | — | 20.9% |
Math Step 5 Preview leads
MiniMax-M2: 37.3 (#160), Step 5 Preview: 46.1 (#72)
| Benchmark | MiniMax-M2 | Step 5 Preview |
|---|---|---|
| LMArena Math | 1352 | 1470 |
| ProofBench | — | 42% |
Knowledge Step 5 Preview leads
MiniMax-M2: 37.0 (#163), Step 5 Preview: 41.2 (#112)
| Benchmark | MiniMax-M2 | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1337 | 1470 |
Multimodal Not comparable
MiniMax-M2: —, Step 5 Preview: 41.0 (#41)
| Benchmark | MiniMax-M2 | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |
Multilingual Step 5 Preview leads
MiniMax-M2: 45.3 (#171), Step 5 Preview: 53.8 (#54)
| Benchmark | MiniMax-M2 | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1313 | 1432 |
| LMArena Chinese | 1366 | 1519 |
| LMArena Russian | 1331 | 1436 |
| LMArena Spanish | 1326 | 1439 |
| LMArena French | 1335 | — |
| LMArena German | 1355 | — |
Instruction Following Step 5 Preview leads
MiniMax-M2: 70.2 (#166), Step 5 Preview: 76.0 (#50)
| Benchmark | MiniMax-M2 | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1328 | 1444 |
Long Context Step 5 Preview leads
MiniMax-M2: 40.5 (#153), Step 5 Preview: 44.5 (#64)
| Benchmark | MiniMax-M2 | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1331 | 1455 |
Writing & Preference Step 5 Preview leads
MiniMax-M2: 53.0 (#162), Step 5 Preview: 62.8 (#71)
| Benchmark | MiniMax-M2 | Step 5 Preview |
|---|---|---|
| LMArena Text | 1340 | 1442 |
| LMArena Creative Writing | 1286 | 1410 |
| LMArena Multi-Turn | 1361 | 1447 |
Frequently asked questions
Is MiniMax-M2 better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 37.4 on the Noometry Index. MiniMax-M2 costs 2.7× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Which is cheaper, MiniMax-M2 or Step 5 Preview?
MiniMax-M2 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Step 5 Preview lists at $1 and $2.70.
Is MiniMax-M2 or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 39.3 in the Noometry coding category.
Which has the bigger context window?
Step 5 Preview does, with 1.02M tokens against 205K.
How many benchmarks do MiniMax-M2 and Step 5 Preview share?
14 benchmarks have published results for both models. MiniMax-M2 has 21 scored results on Noometry and Step 5 Preview has 18.