Model comparison
Ministral 8B vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 28.2 on the Noometry Index. Ministral 8B costs 9.5× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Last verified . 12 shared benchmarks.
Mistral AI
StepFunSummary
- They share 12 benchmarks with published results for both. Ministral 8B scores higher in 0 categories and Step 5 Preview in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Step 5 Preview leads 41.2 to 12.6.
- Ministral 8B is cheaper at $0.15 / $0.15 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
- Step 5 Preview accepts more context: 1.02M tokens versus 262K.
- Ministral 8B has downloadable open weights; the other is API-only.
Side by side
| Ministral 8B | Step 5 Preview | |
|---|---|---|
| Provider | Mistral AI | StepFun |
| Noometry Index | 28.2 | 47.9 |
| Released | 2024-10-01 | 2026-09-16 |
| Weights | Open | Proprietary |
| Context window | 262K | 1.02M |
| Max output | 262K | 66K |
| Input $ / M tokens | $0.15 | $1 |
| Output $ / M tokens | $0.15 | $2.70 |
| Results tracked | 17 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
Ministral 8B: 35.0 (#230), Step 5 Preview: 51.8 (#36)
| Benchmark | Ministral 8B | Step 5 Preview |
|---|---|---|
| LMArena Coding | 1202 | 1480 |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
Agentic & Tool Use Not comparable
Ministral 8B: 16.4 (#148), Step 5 Preview: —
| Benchmark | Ministral 8B | Step 5 Preview |
|---|---|---|
| Berkeley Function Calling Leaderboard | 11.1% | — |
Reasoning Step 5 Preview leads
Ministral 8B: 18.4 (#281), Step 5 Preview: 40.0 (#57)
| Benchmark | Ministral 8B | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1191 | 1465 |
| CritPt | — | 20.9% |
| DTBench | 45.7% | — |
Math Step 5 Preview leads
Ministral 8B: 25.7 (#267), Step 5 Preview: 46.1 (#72)
| Benchmark | Ministral 8B | Step 5 Preview |
|---|---|---|
| LMArena Math | 1188 | 1470 |
| ProofBench | — | 42% |
| MATH Level 5 | 14.9% | — |
Knowledge Step 5 Preview leads
Ministral 8B: 12.6 (#297), Step 5 Preview: 41.2 (#112)
| Benchmark | Ministral 8B | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1170 | 1470 |
| GPQA Diamond | 27.1% | — |
| Vectara Hallucination Rate | 7.4% | — |
Multimodal Not comparable
Ministral 8B: —, Step 5 Preview: 41.0 (#41)
| Benchmark | Ministral 8B | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |
Multilingual Step 5 Preview leads
Ministral 8B: 35.1 (#247), Step 5 Preview: 53.8 (#54)
| Benchmark | Ministral 8B | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1165 | 1432 |
| LMArena Chinese | 1193 | 1519 |
| LMArena Russian | 1195 | 1436 |
| LMArena Spanish | — | 1439 |
Instruction Following Step 5 Preview leads
Ministral 8B: 60.5 (#250), Step 5 Preview: 76.0 (#50)
| Benchmark | Ministral 8B | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1161 | 1444 |
Long Context Step 5 Preview leads
Ministral 8B: 36.7 (#227), Step 5 Preview: 44.5 (#64)
| Benchmark | Ministral 8B | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1212 | 1455 |
Writing & Preference Step 5 Preview leads
Ministral 8B: 39.6 (#246), Step 5 Preview: 62.8 (#71)
| Benchmark | Ministral 8B | Step 5 Preview |
|---|---|---|
| LMArena Text | 1191 | 1442 |
| LMArena Creative Writing | 1175 | 1410 |
| LMArena Multi-Turn | 1166 | 1447 |
Frequently asked questions
Is Ministral 8B better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 28.2 on the Noometry Index. Ministral 8B costs 9.5× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Which is cheaper, Ministral 8B or Step 5 Preview?
Ministral 8B is cheaper. It lists at $0.15 per million input tokens and $0.15 per million output tokens; Step 5 Preview lists at $1 and $2.70.
Is Ministral 8B or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 35.0 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 Ministral 8B and Step 5 Preview share?
12 benchmarks have published results for both models. Ministral 8B has 17 scored results on Noometry and Step 5 Preview has 18.