Model comparison
MiniMax-M2.7 vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 37.7 on the Noometry Index.
Last verified . 15 shared benchmarks.
MiniMax
StepFunSummary
- They share 15 benchmarks with published results for both. MiniMax-M2.7 scores higher in 6 categories and Step 3 in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 3 leads 37.6 to 25.9.
Side by side
| MiniMax-M2.7 | Step 3 | |
|---|---|---|
| Provider | MiniMax | StepFun |
| Noometry Index | 37.7 | 40.5 |
| Released | 2026-03-18 | — |
| Weights | Open | Open |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 30 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.7 leads
MiniMax-M2.7: 41.8 (#120), Step 3: 40.1 (#147)
| Benchmark | MiniMax-M2.7 | Step 3 |
|---|---|---|
| LMArena Coding | 1454 | 1367 |
| LMArena WebDev | 1398 | — |
| SciCode | 47% | — |
| WeirdML | 37% | — |
| ALE-Bench | 599.25 | — |
Agentic & Tool Use Not comparable
MiniMax-M2.7: 25.1 (#111), Step 3: —
| Benchmark | MiniMax-M2.7 | Step 3 |
|---|---|---|
| Terminal-Bench | 45.1% | — |
| ExploitBench | 13.3% | — |
| GBAEval | 0% | — |
Reasoning Step 3 leads
MiniMax-M2.7: 19.7 (#253), Step 3: 28.4 (#105)
| Benchmark | MiniMax-M2.7 | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1422 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
| NYT Connections (extended) | 24.7% | — |
| CritPt | 0.6% | — |
| Thematic Generalization | 39.3% | — |
| Epoch Capabilities Index | 145.85 | — |
Math Step 3 leads
MiniMax-M2.7: 25.9 (#263), Step 3: 37.6 (#148)
| Benchmark | MiniMax-M2.7 | Step 3 |
|---|---|---|
| LMArena Math | 1420 | 1366 |
| ProofBench | 3% | — |
Knowledge Too close to call
MiniMax-M2.7: 37.7 (#152), Step 3: 36.8 (#164)
| Benchmark | MiniMax-M2.7 | Step 3 |
|---|---|---|
| LMArena Expert | 1444 | 1333 |
| Vectara Hallucination Rate | 12.9% | — |
Multimodal Not comparable
MiniMax-M2.7: —, Step 3: 35.5 (#86)
| Benchmark | MiniMax-M2.7 | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual MiniMax-M2.7 leads
MiniMax-M2.7: 50.3 (#123), Step 3: 46.3 (#159)
| Benchmark | MiniMax-M2.7 | Step 3 |
|---|---|---|
| LMArena Non-English | 1382 | 1327 |
| LMArena Chinese | 1441 | 1397 |
| LMArena German | 1398 | 1371 |
| LMArena Korean | 1313 | 1269 |
| LMArena Russian | 1383 | 1331 |
| LMArena Spanish | 1403 | 1371 |
| LMArena French | 1421 | — |
| LMArena Japanese | 1262 | — |
Instruction Following MiniMax-M2.7 leads
MiniMax-M2.7: 74.1 (#103), Step 3: 70.4 (#164)
| Benchmark | MiniMax-M2.7 | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1405 | 1332 |
Long Context MiniMax-M2.7 leads
MiniMax-M2.7: 43.3 (#99), Step 3: 40.3 (#157)
| Benchmark | MiniMax-M2.7 | Step 3 |
|---|---|---|
| LMArena Longer Query | 1419 | 1326 |
Writing & Preference MiniMax-M2.7 leads
MiniMax-M2.7: 58.9 (#112), Step 3: 54.3 (#151)
| Benchmark | MiniMax-M2.7 | Step 3 |
|---|---|---|
| LMArena Text | 1405 | 1350 |
| LMArena Creative Writing | 1354 | 1321 |
| LMArena Multi-Turn | 1412 | 1341 |
Frequently asked questions
Is MiniMax-M2.7 better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 37.7 on the Noometry Index.
Is MiniMax-M2.7 or Step 3 better for coding?
MiniMax-M2.7 scores higher on coding benchmarks: 41.8 versus 40.1 in the Noometry coding category.
How many benchmarks do MiniMax-M2.7 and Step 3 share?
15 benchmarks have published results for both models. MiniMax-M2.7 has 30 scored results on Noometry and Step 3 has 17.