Model comparison
MiniMax-M2.5 vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 43.8 to 38.1 on the Noometry Index. MiniMax-M2.5 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 . 1 shared benchmarks.
MiniMax
StepFunSummary
- They share 1 benchmark with published results for both. MiniMax-M2.5 scores higher in 0 categories and Step 5 Preview in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 5 Preview leads 38.0 to 17.6.
- MiniMax-M2.5 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-M2.5 has downloadable open weights; the other is API-only.
Side by side
| MiniMax-M2.5 | Step 5 Preview | |
|---|---|---|
| Provider | MiniMax | StepFun |
| Noometry Index | 38.1 | 43.8 |
| Released | 2026-02-12 | 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 | 33 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
MiniMax-M2.5: 47.9 (#53), Step 5 Preview: 50.6 (#40)
| Benchmark | MiniMax-M2.5 | Step 5 Preview |
|---|---|---|
| LMArena WebDev | 1386 | 1569 |
| SWE-bench Verified (bash only) | 75.8% | — |
| SWE-bench Multilingual | 68.3% | — |
| SciCode | — | 58.9% |
| LMArena Coding | 1380 | — |
| ALE-Bench | 618.17 | — |
Agentic & Tool Use Not comparable
MiniMax-M2.5: 30.2 (#70), Step 5 Preview: —
| Benchmark | MiniMax-M2.5 | Step 5 Preview |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Vending-Bench 2 | -23.16 | — |
Reasoning Step 5 Preview leads
MiniMax-M2.5: 17.6 (#281), Step 5 Preview: 38.0 (#58)
| Benchmark | MiniMax-M2.5 | Step 5 Preview |
|---|---|---|
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 55.2% | — |
| NYT Connections (extended) | 16.8% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | — | 20.9% |
| LMArena Hard Prompts | 1372 | — |
| Epoch Capabilities Index | 146.68 | — |
Math Not comparable
MiniMax-M2.5: 25.8 (#242), Step 5 Preview: —
| Benchmark | MiniMax-M2.5 | Step 5 Preview |
|---|---|---|
| ProofBench | 4% | — |
| LMArena Math | 1378 | — |
Knowledge Not comparable
MiniMax-M2.5: 38.6 (#132), Step 5 Preview: —
| Benchmark | MiniMax-M2.5 | Step 5 Preview |
|---|---|---|
| Vectara Hallucination Rate | 9.1% | — |
| LMArena Expert | 1378 | — |
Multilingual Not comparable
MiniMax-M2.5: 47.3 (#147), Step 5 Preview: —
| Benchmark | MiniMax-M2.5 | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1338 | — |
| LMArena Chinese | 1392 | — |
| LMArena French | 1363 | — |
| LMArena German | 1361 | — |
| LMArena Japanese | 1169 | — |
| LMArena Korean | 1232 | — |
| LMArena Russian | 1359 | — |
| LMArena Spanish | 1354 | — |
Instruction Following Not comparable
MiniMax-M2.5: 71.7 (#143), Step 5 Preview: —
| Benchmark | MiniMax-M2.5 | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1353 | — |
Long Context Not comparable
MiniMax-M2.5: 37.4 (#211), Step 5 Preview: —
| Benchmark | MiniMax-M2.5 | Step 5 Preview |
|---|---|---|
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |
| LMArena Longer Query | 1367 | — |
Writing & Preference Not comparable
MiniMax-M2.5: 54.1 (#149), Step 5 Preview: —
| Benchmark | MiniMax-M2.5 | Step 5 Preview |
|---|---|---|
| LMArena Text | 1359 | — |
| LMArena Creative Writing | 1331 | — |
| EQ-Bench Creative Writing | 1361 | — |
| LMArena Multi-Turn | 1364 | — |
Frequently asked questions
Is MiniMax-M2.5 better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 43.8 to 38.1 on the Noometry Index. MiniMax-M2.5 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-M2.5 or Step 5 Preview?
MiniMax-M2.5 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-M2.5 or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 50.6 versus 47.9 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-M2.5 and Step 5 Preview share?
1 benchmark has published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Step 5 Preview has 3.