Model comparison
MiniMax-M2.7 vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 37.7 on the Noometry Index. MiniMax-M2.7 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 . 17 shared benchmarks.
MiniMax
StepFunSummary
- They share 17 benchmarks with published results for both. MiniMax-M2.7 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 19.7.
- The biggest single-benchmark swing is ProofBench: 3% for MiniMax-M2.7 and 42% for Step 5 Preview.
- MiniMax-M2.7 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-M2.7 has downloadable open weights; the other is API-only.
Side by side
| MiniMax-M2.7 | Step 5 Preview | |
|---|---|---|
| Provider | MiniMax | StepFun |
| Noometry Index | 37.7 | 47.9 |
| Released | 2026-03-18 | 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 | 30 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
MiniMax-M2.7: 41.8 (#120), Step 5 Preview: 51.8 (#36)
| Benchmark | MiniMax-M2.7 | Step 5 Preview |
|---|---|---|
| LMArena WebDev | 1398 | 1564 |
| SciCode | 47% | 58.9% |
| LMArena Coding | 1454 | 1480 |
| WeirdML | 37% | — |
| ALE-Bench | 599.25 | — |
Agentic & Tool Use Not comparable
MiniMax-M2.7: 25.1 (#111), Step 5 Preview: —
| Benchmark | MiniMax-M2.7 | Step 5 Preview |
|---|---|---|
| Terminal-Bench | 45.1% | — |
| ExploitBench | 13.3% | — |
| GBAEval | 0% | — |
Reasoning Step 5 Preview leads
MiniMax-M2.7: 19.7 (#253), Step 5 Preview: 40.0 (#57)
| Benchmark | MiniMax-M2.7 | Step 5 Preview |
|---|---|---|
| CritPt | 0.6% | 20.9% |
| LMArena Hard Prompts | 1422 | 1465 |
| NYT Connections (extended) | 24.7% | — |
| Thematic Generalization | 39.3% | — |
| Epoch Capabilities Index | 145.85 | — |
Math Step 5 Preview leads
MiniMax-M2.7: 25.9 (#263), Step 5 Preview: 46.1 (#72)
| Benchmark | MiniMax-M2.7 | Step 5 Preview |
|---|---|---|
| ProofBench | 3% | 42% |
| LMArena Math | 1420 | 1470 |
Knowledge Step 5 Preview leads
MiniMax-M2.7: 37.7 (#152), Step 5 Preview: 41.2 (#112)
| Benchmark | MiniMax-M2.7 | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1444 | 1470 |
| Vectara Hallucination Rate | 12.9% | — |
Multimodal Not comparable
MiniMax-M2.7: —, Step 5 Preview: 41.0 (#41)
| Benchmark | MiniMax-M2.7 | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |
Multilingual Step 5 Preview leads
MiniMax-M2.7: 50.3 (#123), Step 5 Preview: 53.8 (#54)
| Benchmark | MiniMax-M2.7 | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1382 | 1432 |
| LMArena Chinese | 1441 | 1519 |
| LMArena Russian | 1383 | 1436 |
| LMArena Spanish | 1403 | 1439 |
| LMArena French | 1421 | — |
| LMArena German | 1398 | — |
| LMArena Japanese | 1262 | — |
| LMArena Korean | 1313 | — |
Instruction Following Step 5 Preview leads
MiniMax-M2.7: 74.1 (#103), Step 5 Preview: 76.0 (#50)
| Benchmark | MiniMax-M2.7 | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1405 | 1444 |
Long Context Step 5 Preview leads
MiniMax-M2.7: 43.3 (#99), Step 5 Preview: 44.5 (#64)
| Benchmark | MiniMax-M2.7 | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1419 | 1455 |
Writing & Preference Step 5 Preview leads
MiniMax-M2.7: 58.9 (#112), Step 5 Preview: 62.8 (#71)
| Benchmark | MiniMax-M2.7 | Step 5 Preview |
|---|---|---|
| LMArena Text | 1405 | 1442 |
| LMArena Creative Writing | 1354 | 1410 |
| LMArena Multi-Turn | 1412 | 1447 |
Frequently asked questions
Is MiniMax-M2.7 better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 37.7 on the Noometry Index. MiniMax-M2.7 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-M2.7 or Step 5 Preview?
MiniMax-M2.7 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-M2.7 or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 41.8 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-M2.7 and Step 5 Preview share?
17 benchmarks have published results for both models. MiniMax-M2.7 has 30 scored results on Noometry and Step 5 Preview has 18.