# Mistral Small vs Step 5 Preview

> Step 5 Preview is the stronger model overall, scoring 47.9 to 33.4 on the Noometry Index. Mistral Small costs 5.4× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/mistral-small-vs-step-5-preview
- Last updated: 2026-10-11
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. Mistral Small scores higher in 0 categories and Step 5 Preview in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 5 Preview leads 46.1 to 16.4.
- The biggest single-benchmark swing is SciCode: 26.5% for Mistral Small and 58.9% for Step 5 Preview.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
- Step 5 Preview accepts more context: 1.02M tokens versus 262K.
- Mistral Small has downloadable open weights; the other is API-only.

## Snapshot

| | Mistral Small | Step 5 Preview |
|---|---|---|
| Provider | Mistral AI | StepFun |
| Noometry Index | 33.4 | 47.9 |
| Rank | 243 | 58 |
| Context | 262K | 1.02M |
| Input $/M | $0.15 | $1 |
| Output $/M | $0.60 | $2.70 |
| Weights | Open | Proprietary |

## Coding

- Mistral Small: 34.0 (#247)
- Step 5 Preview: 51.8 (#36)

| Benchmark | Mistral Small | Step 5 Preview |
|---|---|---|
| SciCode | 26.5% | 58.9% |
| LMArena Coding | 1362 | 1480 |
| LMArena WebDev | — | 1564 |
| BigCodeBench Instruct | 36.1% | — |
| LiveBench Coding | 36.2% | — |
| BigCodeBench Complete | 46.6% | — |
| ALE-Bench | 497.62 | — |

## Agentic & Tool Use

- Mistral Small: 28.1 (#93)
- Step 5 Preview: —

| Benchmark | Mistral Small | Step 5 Preview |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.1% | — |

## Reasoning

- Mistral Small: 19.8 (#250)
- Step 5 Preview: 40.0 (#57)

| Benchmark | Mistral Small | Step 5 Preview |
|---|---|---|
| CritPt | 0% | 20.9% |
| LMArena Hard Prompts | 1335 | 1465 |
| Kagi LLM Benchmark | 37.8% | — |
| LiveBench Reasoning | 44.8% | — |
| DTBench | 70.9% | — |
| LiveBench Data Analysis | 53.7% | — |
| LMCA | 20.6% | — |
| LiveBench | 44% | — |

## Math

- Mistral Small: 16.4 (#293)
- Step 5 Preview: 46.1 (#72)

| Benchmark | Mistral Small | Step 5 Preview |
|---|---|---|
| LMArena Math | 1341 | 1470 |
| OTIS Mock AIME 2024-2025 | 5.8% | — |
| ProofBench | — | 42% |
| LiveBench Math | 39.9% | — |
| MATH Level 5 | 46.8% | — |

## Knowledge

- Mistral Small: 31.0 (#222)
- Step 5 Preview: 41.2 (#112)

| Benchmark | Mistral Small | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1291 | 1470 |
| GPQA Diamond | 47.5% | — |
| Vectara Hallucination Rate | 5.1% | — |
| MMLU | 68.7% | — |

## Multimodal

- Mistral Small: 33.5 (#96)
- Step 5 Preview: 41.0 (#41)

| Benchmark | Mistral Small | Step 5 Preview |
|---|---|---|
| LMArena Vision | 1142 | 1267 |

## Multilingual

- Mistral Small: 45.5 (#169)
- Step 5 Preview: 53.8 (#54)

| Benchmark | Mistral Small | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1315 | 1432 |
| LMArena Chinese | 1340 | 1519 |
| LMArena Russian | 1324 | 1436 |
| LMArena Spanish | 1346 | 1439 |
| LMArena French | 1337 | — |
| LMArena German | 1340 | — |
| LMArena Japanese | 1275 | — |
| LMArena Korean | 1259 | — |

## Instruction Following

- Mistral Small: 66.4 (#209)
- Step 5 Preview: 76.0 (#50)

| Benchmark | Mistral Small | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1310 | 1444 |
| LiveBench Instruction Following | 63.7% | — |

## Long Context

- Mistral Small: 40.4 (#156)
- Step 5 Preview: 44.5 (#64)

| Benchmark | Mistral Small | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1327 | 1455 |

## Writing & Preference

- Mistral Small: 52.5 (#171)
- Step 5 Preview: 62.8 (#71)

| Benchmark | Mistral Small | Step 5 Preview |
|---|---|---|
| LMArena Text | 1338 | 1442 |
| LMArena Creative Writing | 1305 | 1410 |
| LMArena Multi-Turn | 1344 | 1447 |
| LiveBench Language | 30.5% | — |

## FAQ

### Is Mistral Small better than Step 5 Preview?

Step 5 Preview is the stronger model overall, scoring 47.9 to 33.4 on the Noometry Index. Mistral Small costs 5.4× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.

### Which is cheaper, Mistral Small or Step 5 Preview?

Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Step 5 Preview lists at $1 and $2.70.

### Is Mistral Small or Step 5 Preview better for coding?

Step 5 Preview scores higher on coding benchmarks: 51.8 versus 34.0 in the Noometry coding category.

### Which has the bigger context window?

Step 5 Preview does, with 1.02M tokens against 262K.

### How many benchmarks do Mistral Small and Step 5 Preview share?

16 benchmarks have published results for both models. Mistral Small has 39 scored results on Noometry and Step 5 Preview has 18.