Model comparison
MiniMax-M2.1 vs Step 3.7 Flash
MiniMax-M2.1 is the stronger model overall, scoring 38.9 to 37.3 on the Noometry Index.
Last verified . 2 shared benchmarks.
MiniMax
StepFunSummary
- They share 2 benchmarks with published results for both. MiniMax-M2.1 scores higher in 1 category and Step 3.7 Flash in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 3.7 Flash leads 21.6 to 16.6.
- The biggest single-benchmark swing is NYT Connections (extended): 11.2% for MiniMax-M2.1 and 39.7% for Step 3.7 Flash.
- Step 3.7 Flash is cheaper at $0.18 / $1.11 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.1.
- Step 3.7 Flash accepts more context: 256K tokens versus 205K.
Side by side
| MiniMax-M2.1 | Step 3.7 Flash | |
|---|---|---|
| Provider | MiniMax | StepFun |
| Noometry Index | 38.9 | 37.3 |
| Released | 2025-12-23 | 2026-05-29 |
| Weights | Open | Open |
| Context window | 205K | 256K |
| Max output | 131K | 256K |
| Input $ / M tokens | $0.30 | $0.18 |
| Output $ / M tokens | $1.20 | $1.11 |
| Results tracked | 22 | 5 |
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 3.7 Flash: 40.0 (#150)
| Benchmark | MiniMax-M2.1 | Step 3.7 Flash |
|---|---|---|
| ALE-Bench | 623.83 | 694.12 |
| LMArena WebDev | 1384 | — |
| SciCode | — | 40% |
| LMArena Coding | 1421 | — |
Agentic & Tool Use Not comparable
MiniMax-M2.1: 27.9 (#98), Step 3.7 Flash: —
| Benchmark | MiniMax-M2.1 | Step 3.7 Flash |
|---|---|---|
| Terminal-Bench | 36.6% | — |
Reasoning Step 3.7 Flash leads
MiniMax-M2.1: 16.6 (#302), Step 3.7 Flash: 21.6 (#219)
| Benchmark | MiniMax-M2.1 | Step 3.7 Flash |
|---|---|---|
| NYT Connections (extended) | 11.2% | 39.7% |
| CritPt | — | 2.3% |
| LMArena Hard Prompts | 1411 | — |
Math Step 3.7 Flash leads
MiniMax-M2.1: 38.3 (#138), Step 3.7 Flash: 42.9 (#82)
| Benchmark | MiniMax-M2.1 | Step 3.7 Flash |
|---|---|---|
| MathArena Final-Answer Competitions | — | 68.5% |
| LMArena Math | 1397 | — |
Knowledge Not comparable
MiniMax-M2.1: 38.3 (#147), Step 3.7 Flash: —
| Benchmark | MiniMax-M2.1 | Step 3.7 Flash |
|---|---|---|
| Vectara Hallucination Rate | 11.8% | — |
| LMArena Expert | 1431 | — |
Multilingual Not comparable
MiniMax-M2.1: 50.0 (#128), Step 3.7 Flash: —
| Benchmark | MiniMax-M2.1 | Step 3.7 Flash |
|---|---|---|
| LMArena Non-English | 1378 | — |
| LMArena Chinese | 1430 | — |
| LMArena French | 1404 | — |
| LMArena German | 1381 | — |
| LMArena Japanese | 1287 | — |
| LMArena Korean | 1298 | — |
| LMArena Russian | 1387 | — |
| LMArena Spanish | 1397 | — |
Instruction Following Not comparable
MiniMax-M2.1: 73.8 (#112), Step 3.7 Flash: —
| Benchmark | MiniMax-M2.1 | Step 3.7 Flash |
|---|---|---|
| LMArena Instruction Following | 1400 | — |
Long Context Not comparable
MiniMax-M2.1: 43.2 (#101), Step 3.7 Flash: —
| Benchmark | MiniMax-M2.1 | Step 3.7 Flash |
|---|---|---|
| LMArena Longer Query | 1416 | — |
Writing & Preference Not comparable
MiniMax-M2.1: 58.3 (#120), Step 3.7 Flash: —
| Benchmark | MiniMax-M2.1 | Step 3.7 Flash |
|---|---|---|
| LMArena Text | 1392 | — |
| LMArena Creative Writing | 1361 | — |
| LMArena Multi-Turn | 1396 | — |
Frequently asked questions
Is MiniMax-M2.1 better than Step 3.7 Flash?
MiniMax-M2.1 is the stronger model overall, scoring 38.9 to 37.3 on the Noometry Index.
Which is cheaper, MiniMax-M2.1 or Step 3.7 Flash?
Step 3.7 Flash is cheaper. It lists at $0.18 per million input tokens and $1.11 per million output tokens; MiniMax-M2.1 lists at $0.30 and $1.20.
Is MiniMax-M2.1 or Step 3.7 Flash better for coding?
They score almost the same on coding (40.4 vs 40.0); test both on your own repository before choosing.
Which has the bigger context window?
Step 3.7 Flash does, with 256K tokens against 205K.
How many benchmarks do MiniMax-M2.1 and Step 3.7 Flash share?
2 benchmarks have published results for both models. MiniMax-M2.1 has 22 scored results on Noometry and Step 3.7 Flash has 5.