Model comparison
Ministral 8B vs Mistral Large
Mistral Large is the stronger model overall, scoring 31.9 to 28.2 on the Noometry Index. Ministral 8B costs 20× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
Mistral AISummary
- They share 17 benchmarks with published results for both. Ministral 8B scores higher in 3 categories and Mistral Large in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mistral Large leads 30.1 to 12.6.
- The biggest single-benchmark swing is MATH Level 5: 14.9% for Ministral 8B and 50.3% for Mistral Large.
- Ministral 8B is cheaper at $0.15 / $0.15 per million input/output tokens, against $2 / $6 for Mistral Large.
- Ministral 8B accepts more context: 262K tokens versus 131K.
Side by side
| Ministral 8B | Mistral Large | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 28.2 | 31.9 |
| Released | 2024-10-01 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 262K | 16K |
| Input $ / M tokens | $0.15 | $2 |
| Output $ / M tokens | $0.15 | $6 |
| Results tracked | 17 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Ministral 8B: 35.0 (#230), Mistral Large: 34.3 (#240)
| Benchmark | Ministral 8B | Mistral Large |
|---|---|---|
| LMArena Coding | 1202 | 1277 |
| 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 Mistral Large leads
Ministral 8B: 16.4 (#148), Mistral Large: 28.6 (#89)
| Benchmark | Ministral 8B | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | 11.1% | 38.4% |
Reasoning Ministral 8B leads
Ministral 8B: 18.4 (#281), Mistral Large: 15.8 (#310)
| Benchmark | Ministral 8B | Mistral Large |
|---|---|---|
| LMArena Hard Prompts | 1191 | 1257 |
| DTBench | 45.7% | 65.1% |
| SimpleBench | — | 22.5% |
| CritPt | — | 0% |
| LiveBench Reasoning | — | 43.5% |
| LiveBench Data Analysis | — | 50.1% |
| LMCA | — | 16.7% |
| Epoch Capabilities Index | — | 128.52 |
| ForecastBench | — | 57.1 |
| LiveBench | — | 48.4% |
Math Ministral 8B leads
Ministral 8B: 25.7 (#267), Mistral Large: 18.2 (#291)
| Benchmark | Ministral 8B | Mistral Large |
|---|---|---|
| LMArena Math | 1188 | 1262 |
| MATH Level 5 | 14.9% | 50.3% |
| OTIS Mock AIME 2024-2025 | — | 8.5% |
| Omni-MATH | — | 28.1% |
| LiveBench Math | — | 42.5% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Mistral Large leads
Ministral 8B: 12.6 (#297), Mistral Large: 30.1 (#230)
| Benchmark | Ministral 8B | Mistral Large |
|---|---|---|
| GPQA Diamond | 27.1% | 51.3% |
| Vectara Hallucination Rate | 7.4% | 4.5% |
| LMArena Expert | 1170 | 1232 |
| MMLU-Pro | — | 59.9% |
| Confabulations | — | 21.4% |
| GPQA (HELM) | — | 43.5% |
| MMLU | — | 80% |
Multilingual Mistral Large leads
Ministral 8B: 35.1 (#247), Mistral Large: 40.0 (#219)
| Benchmark | Ministral 8B | Mistral Large |
|---|---|---|
| LMArena Non-English | 1165 | 1237 |
| LMArena Chinese | 1193 | 1240 |
| LMArena Russian | 1195 | 1257 |
| LMArena French | — | 1325 |
| LMArena German | — | 1254 |
| LMArena Japanese | — | 1188 |
| LMArena Korean | — | 1202 |
| LMArena Spanish | — | 1268 |
Instruction Following Mistral Large leads
Ministral 8B: 60.5 (#250), Mistral Large: 67.9 (#191)
| Benchmark | Ministral 8B | Mistral Large |
|---|---|---|
| LMArena Instruction Following | 1161 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
| IFEval | — | 87.7% |
Long Context Mistral Large leads
Ministral 8B: 36.7 (#227), Mistral Large: 38.3 (#199)
| Benchmark | Ministral 8B | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1212 | 1261 |
Writing & Preference Mistral Large leads
Ministral 8B: 39.6 (#246), Mistral Large: 40.7 (#242)
| Benchmark | Ministral 8B | Mistral Large |
|---|---|---|
| LMArena Text | 1191 | 1266 |
| LMArena Creative Writing | 1175 | 1243 |
| LMArena Multi-Turn | 1166 | 1260 |
| Short-Story Creative Writing | — | 69% |
| EQ-Bench Creative Writing | — | 985 |
| WildBench | — | 80.1% |
| LiveBench Language | — | 39.4% |
Frequently asked questions
Is Ministral 8B better than Mistral Large?
Mistral Large is the stronger model overall, scoring 31.9 to 28.2 on the Noometry Index. Ministral 8B costs 20× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Which is cheaper, Ministral 8B or Mistral Large?
Ministral 8B is cheaper. It lists at $0.15 per million input tokens and $0.15 per million output tokens; Mistral Large lists at $2 and $6.
Is Ministral 8B or Mistral Large better for coding?
They score almost the same on coding (35.0 vs 34.3); test both on your own repository before choosing.
Which has the bigger context window?
Ministral 8B does, with 262K tokens against 131K.
How many benchmarks do Ministral 8B and Mistral Large share?
17 benchmarks have published results for both models. Ministral 8B has 17 scored results on Noometry and Mistral Large has 51.