Model comparison
MiMo-V2.6-Pro vs Step 3
MiMo-V2.6-Pro is the stronger model overall, scoring 50.3 to 40.5 on the Noometry Index.
Last verified . 13 shared benchmarks.
StepFunSummary
- They share 13 benchmarks with published results for both. MiMo-V2.6-Pro scores higher in 9 categories and Step 3 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where MiMo-V2.6-Pro leads 54.5 to 37.6.
Side by side
| MiMo-V2.6-Pro | Step 3 | |
|---|---|---|
| Provider | Xiaomi | StepFun |
| Noometry Index | 50.3 | 40.5 |
| Released | 2026-09-21 | — |
| Weights | Open | Open |
| Context window | 1.05M | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.43 | — |
| Output $ / M tokens | $0.87 | — |
| Results tracked | 19 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 55.5 (#23), Step 3: 40.1 (#147)
| Benchmark | MiMo-V2.6-Pro | Step 3 |
|---|---|---|
| LMArena Coding | 1534 | 1367 |
| LMArena WebDev | 1629 | — |
| SciCode | 60.9% | — |
| ALE-Bench | 1,158 | — |
Agentic & Tool Use Not comparable
MiMo-V2.6-Pro: 37.5 (#35), Step 3: —
| Benchmark | MiMo-V2.6-Pro | Step 3 |
|---|---|---|
| APEX-Agents | 59.5% | — |
Reasoning MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 43.1 (#50), Step 3: 28.4 (#105)
| Benchmark | MiMo-V2.6-Pro | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1512 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
| CritPt | 26.6% | — |
Math MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 54.5 (#45), Step 3: 37.6 (#148)
| Benchmark | MiMo-V2.6-Pro | Step 3 |
|---|---|---|
| LMArena Math | 1494 | 1366 |
| ProofBench | 70% | — |
Knowledge MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 43.5 (#92), Step 3: 36.8 (#164)
| Benchmark | MiMo-V2.6-Pro | Step 3 |
|---|---|---|
| LMArena Expert | 1543 | 1333 |
Multimodal MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 40.8 (#43), Step 3: 35.5 (#86)
| Benchmark | MiMo-V2.6-Pro | Step 3 |
|---|---|---|
| LMArena Vision | 1264 | 1177 |
Multilingual MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 56.9 (#14), Step 3: 46.3 (#159)
| Benchmark | MiMo-V2.6-Pro | Step 3 |
|---|---|---|
| LMArena Non-English | 1474 | 1327 |
| LMArena Chinese | 1529 | 1397 |
| LMArena Russian | 1480 | 1331 |
| LMArena German | — | 1371 |
| LMArena Korean | — | 1269 |
| LMArena Spanish | — | 1371 |
Instruction Following MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 78.2 (#12), Step 3: 70.4 (#164)
| Benchmark | MiMo-V2.6-Pro | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1493 | 1332 |
Long Context MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 46.0 (#27), Step 3: 40.3 (#157)
| Benchmark | MiMo-V2.6-Pro | Step 3 |
|---|---|---|
| LMArena Longer Query | 1501 | 1326 |
Writing & Preference MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 66.8 (#33), Step 3: 54.3 (#151)
| Benchmark | MiMo-V2.6-Pro | Step 3 |
|---|---|---|
| LMArena Text | 1492 | 1350 |
| LMArena Creative Writing | 1468 | 1321 |
| LMArena Multi-Turn | 1464 | 1341 |
Frequently asked questions
Is MiMo-V2.6-Pro better than Step 3?
MiMo-V2.6-Pro is the stronger model overall, scoring 50.3 to 40.5 on the Noometry Index.
Is MiMo-V2.6-Pro or Step 3 better for coding?
MiMo-V2.6-Pro scores higher on coding benchmarks: 55.5 versus 40.1 in the Noometry coding category.
How many benchmarks do MiMo-V2.6-Pro and Step 3 share?
13 benchmarks have published results for both models. MiMo-V2.6-Pro has 19 scored results on Noometry and Step 3 has 17.