Model comparison
Mistral vs Step 3
Step 3 is the stronger model overall, scoring 39.9 to 29.4 on the Noometry Index.
Last verified . 15 shared benchmarks.
Mistral AI
StepFunSummary
- They share 15 benchmarks with published results for both. Mistral scores higher in 0 categories and Step 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Step 3 leads 36.2 to 16.3.
- Step 3 has downloadable open weights; the other is API-only.
Side by side
| Mistral | Step 3 | |
|---|---|---|
| Provider | Mistral AI | StepFun |
| Noometry Index | 29.4 | 39.9 |
| Released | — | — |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Mistral: 33.2 (#249), Step 3: 39.5 (#144)
| Benchmark | Mistral | Step 3 |
|---|---|---|
| LMArena Coding | 1162 | 1367 |
Reasoning Step 3 leads
Mistral: 21.3 (#199), Step 3: 27.6 (#102)
| Benchmark | Mistral | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1149 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
Math Step 3 leads
Mistral: 20.5 (#275), Step 3: 34.9 (#148)
| Benchmark | Mistral | Step 3 |
|---|---|---|
| LMArena Math | 1180 | 1366 |
| Omni-MATH | 7.2% | — |
Knowledge Step 3 leads
Mistral: 16.3 (#283), Step 3: 36.2 (#161)
| Benchmark | Mistral | Step 3 |
|---|---|---|
| LMArena Expert | 1125 | 1333 |
| MMLU-Pro | 27.7% | — |
| GPQA (HELM) | 30.3% | — |
Multimodal Not comparable
Mistral: —, Step 3: 34.9 (#81)
| Benchmark | Mistral | Step 3 |
|---|---|---|
| LMArena Vision | — | 1176 |
Multilingual Step 3 leads
Mistral: 33.0 (#249), Step 3: 46.5 (#154)
| Benchmark | Mistral | Step 3 |
|---|---|---|
| LMArena Non-English | 1129 | 1327 |
| LMArena Chinese | 1109 | 1396 |
| LMArena German | 1155 | 1371 |
| LMArena Korean | 1032 | 1271 |
| LMArena Russian | 1168 | 1331 |
| LMArena Spanish | 1143 | 1372 |
| LMArena French | 1179 | — |
| LMArena Japanese | 1012 | — |
Instruction Following Step 3 leads
Mistral: 52.9 (#283), Step 3: 70.6 (#159)
| Benchmark | Mistral | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1152 | 1332 |
| IFEval | 56.8% | — |
Long Context Step 3 leads
Mistral: 35.0 (#240), Step 3: 40.3 (#152)
| Benchmark | Mistral | Step 3 |
|---|---|---|
| LMArena Longer Query | 1154 | 1326 |
Writing & Preference Step 3 leads
Mistral: 37.2 (#255), Step 3: 54.5 (#146)
| Benchmark | Mistral | Step 3 |
|---|---|---|
| LMArena Text | 1165 | 1351 |
| LMArena Creative Writing | 1158 | 1322 |
| LMArena Multi-Turn | 1147 | 1340 |
| WildBench | 66% | — |
Frequently asked questions
Is Mistral better than Step 3?
Step 3 is the stronger model overall, scoring 39.9 to 29.4 on the Noometry Index.
Is Mistral or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 39.5 versus 33.2 in the Noometry coding category.
How many benchmarks do Mistral and Step 3 share?
15 benchmarks have published results for both models. Mistral has 22 scored results on Noometry and Step 3 has 17.