Model comparison
Mistral Large vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 31.9 on the Noometry Index.
Last verified . 15 shared benchmarks.
Mistral AI
StepFunSummary
- They share 15 benchmarks with published results for both. Mistral Large scores higher in 0 categories and Step 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 3 leads 37.6 to 18.2.
Side by side
| Mistral Large | Step 3 | |
|---|---|---|
| Provider | Mistral AI | StepFun |
| Noometry Index | 31.9 | 40.5 |
| Released | 2024-02-26 | — |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 16K | — |
| Input $ / M tokens | $2 | — |
| Output $ / M tokens | $6 | — |
| Results tracked | 51 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Mistral Large: 34.3 (#240), Step 3: 40.1 (#147)
| Benchmark | Mistral Large | Step 3 |
|---|---|---|
| LMArena Coding | 1277 | 1367 |
| SciCode | 36.2% | — |
| BigCodeBench Instruct | 30% | — |
| LiveBench Coding | 47.1% | — |
| BigCodeBench Complete | 38.3% | — |
| ALE-Bench | 264.7 | — |
| HumanEval+ | 62.2% | — |
| MBPP+ | 59.5% | — |
Agentic & Tool Use Not comparable
Mistral Large: 28.6 (#89), Step 3: —
| Benchmark | Mistral Large | Step 3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 38.4% | — |
Reasoning Step 3 leads
Mistral Large: 15.8 (#310), Step 3: 28.4 (#105)
| Benchmark | Mistral Large | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1257 | 1355 |
| SimpleBench | 22.5% | — |
| Kagi LLM Benchmark | — | 62.3% |
| CritPt | 0% | — |
| LiveBench Reasoning | 43.5% | — |
| DTBench | 65.1% | — |
| LiveBench Data Analysis | 50.1% | — |
| LMCA | 16.7% | — |
| Epoch Capabilities Index | 128.52 | — |
| ForecastBench | 57.1 | — |
| LiveBench | 48.4% | — |
Math Step 3 leads
Mistral Large: 18.2 (#291), Step 3: 37.6 (#148)
| Benchmark | Mistral Large | Step 3 |
|---|---|---|
| LMArena Math | 1262 | 1366 |
| OTIS Mock AIME 2024-2025 | 8.5% | — |
| Omni-MATH | 28.1% | — |
| LiveBench Math | 42.5% | — |
| MATH Level 5 | 50.3% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Step 3 leads
Mistral Large: 30.1 (#230), Step 3: 36.8 (#164)
| Benchmark | Mistral Large | Step 3 |
|---|---|---|
| LMArena Expert | 1232 | 1333 |
| GPQA Diamond | 51.3% | — |
| MMLU-Pro | 59.9% | — |
| Confabulations | 21.4% | — |
| Vectara Hallucination Rate | 4.5% | — |
| GPQA (HELM) | 43.5% | — |
| MMLU | 80% | — |
Multimodal Not comparable
Mistral Large: —, Step 3: 35.5 (#86)
| Benchmark | Mistral Large | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual Step 3 leads
Mistral Large: 40.0 (#219), Step 3: 46.3 (#159)
| Benchmark | Mistral Large | Step 3 |
|---|---|---|
| LMArena Non-English | 1237 | 1327 |
| LMArena Chinese | 1240 | 1397 |
| LMArena German | 1254 | 1371 |
| LMArena Korean | 1202 | 1269 |
| LMArena Russian | 1257 | 1331 |
| LMArena Spanish | 1268 | 1371 |
| LMArena French | 1325 | — |
| LMArena Japanese | 1188 | — |
Instruction Following Step 3 leads
Mistral Large: 67.9 (#191), Step 3: 70.4 (#164)
| Benchmark | Mistral Large | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1249 | 1332 |
| LiveBench Instruction Following | 67.9% | — |
| IFEval | 87.7% | — |
Long Context Step 3 leads
Mistral Large: 38.3 (#199), Step 3: 40.3 (#157)
| Benchmark | Mistral Large | Step 3 |
|---|---|---|
| LMArena Longer Query | 1261 | 1326 |
Writing & Preference Step 3 leads
Mistral Large: 40.7 (#242), Step 3: 54.3 (#151)
| Benchmark | Mistral Large | Step 3 |
|---|---|---|
| LMArena Text | 1266 | 1350 |
| LMArena Creative Writing | 1243 | 1321 |
| LMArena Multi-Turn | 1260 | 1341 |
| Short-Story Creative Writing | 69% | — |
| EQ-Bench Creative Writing | 985 | — |
| WildBench | 80.1% | — |
| LiveBench Language | 39.4% | — |
Frequently asked questions
Is Mistral Large better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 31.9 on the Noometry Index.
Is Mistral Large or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 34.3 in the Noometry coding category.
How many benchmarks do Mistral Large and Step 3 share?
15 benchmarks have published results for both models. Mistral Large has 51 scored results on Noometry and Step 3 has 17.