Model comparison
Mistral vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 29.9 on the Noometry Index.
Last verified . 13 shared benchmarks.
Mistral AI
StepFunSummary
- They share 13 benchmarks with published results for both. Mistral scores higher in 0 categories and Step 5 Preview in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Step 5 Preview leads 62.8 to 37.0.
Side by side
| Mistral | Step 5 Preview | |
|---|---|---|
| Provider | Mistral AI | StepFun |
| Noometry Index | 29.9 | 47.9 |
| Released | — | 2026-09-16 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.02M |
| Max output | — | 66K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $2.70 |
| Results tracked | 22 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
Mistral: 33.8 (#250), Step 5 Preview: 51.8 (#36)
| Benchmark | Mistral | Step 5 Preview |
|---|---|---|
| LMArena Coding | 1162 | 1480 |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
Reasoning Step 5 Preview leads
Mistral: 22.2 (#200), Step 5 Preview: 40.0 (#57)
| Benchmark | Mistral | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1149 | 1465 |
| CritPt | — | 20.9% |
Math Step 5 Preview leads
Mistral: 22.3 (#278), Step 5 Preview: 46.1 (#72)
| Benchmark | Mistral | Step 5 Preview |
|---|---|---|
| LMArena Math | 1180 | 1470 |
| ProofBench | — | 42% |
| Omni-MATH | 7.2% | — |
Knowledge Step 5 Preview leads
Mistral: 16.6 (#288), Step 5 Preview: 41.2 (#112)
| Benchmark | Mistral | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1125 | 1470 |
| MMLU-Pro | 27.7% | — |
| GPQA (HELM) | 30.3% | — |
Multimodal Not comparable
Mistral: —, Step 5 Preview: 41.0 (#41)
| Benchmark | Mistral | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |
Multilingual Step 5 Preview leads
Mistral: 32.8 (#254), Step 5 Preview: 53.8 (#54)
| Benchmark | Mistral | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1129 | 1432 |
| LMArena Chinese | 1109 | 1519 |
| LMArena Russian | 1168 | 1436 |
| LMArena Spanish | 1143 | 1439 |
| LMArena French | 1180 | — |
| LMArena German | 1155 | — |
| LMArena Japanese | 1013 | — |
| LMArena Korean | 1032 | — |
Instruction Following Step 5 Preview leads
Mistral: 52.6 (#288), Step 5 Preview: 76.0 (#50)
| Benchmark | Mistral | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1152 | 1444 |
| IFEval | 56.8% | — |
Long Context Step 5 Preview leads
Mistral: 35.0 (#245), Step 5 Preview: 44.5 (#64)
| Benchmark | Mistral | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1153 | 1455 |
Writing & Preference Step 5 Preview leads
Mistral: 37.0 (#260), Step 5 Preview: 62.8 (#71)
| Benchmark | Mistral | Step 5 Preview |
|---|---|---|
| LMArena Text | 1165 | 1442 |
| LMArena Creative Writing | 1158 | 1410 |
| LMArena Multi-Turn | 1147 | 1447 |
| WildBench | 66% | — |
Frequently asked questions
Is Mistral better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 29.9 on the Noometry Index.
Is Mistral or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 33.8 in the Noometry coding category.
How many benchmarks do Mistral and Step 5 Preview share?
13 benchmarks have published results for both models. Mistral has 22 scored results on Noometry and Step 5 Preview has 18.