Model comparison
Mistral Large vs Mistral Small
Mistral Small is the stronger model overall, scoring 33.4 to 31.9 on the Noometry Index.
Last verified . 37 shared benchmarks.
Mistral AISummary
- They share 37 benchmarks with published results for both. Mistral Large scores higher in 4 categories and Mistral Small in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Small leads 52.5 to 40.7.
- The biggest single-benchmark swing is LiveBench Coding: 47.1% for Mistral Large and 36.2% for Mistral Small.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $2 / $6 for Mistral Large.
- Mistral Small accepts more context: 262K tokens versus 131K.
Side by side
| Mistral Large | Mistral Small | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 31.9 | 33.4 |
| Released | 2024-02-26 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 16K | 256K |
| Input $ / M tokens | $2 | $0.15 |
| Output $ / M tokens | $6 | $0.60 |
| Results tracked | 51 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Mistral Large: 34.3 (#240), Mistral Small: 34.0 (#247)
| Benchmark | Mistral Large | Mistral Small |
|---|---|---|
| SciCode | 36.2% | 26.5% |
| BigCodeBench Instruct | 30% | 36.1% |
| LiveBench Coding | 47.1% | 36.2% |
| LMArena Coding | 1277 | 1362 |
| BigCodeBench Complete | 38.3% | 46.6% |
| ALE-Bench | 264.7 | 497.62 |
| HumanEval+ | 62.2% | — |
| MBPP+ | 59.5% | — |
Agentic & Tool Use Too close to call
Mistral Large: 28.6 (#89), Mistral Small: 28.1 (#93)
| Benchmark | Mistral Large | Mistral Small |
|---|---|---|
| Berkeley Function Calling Leaderboard | 38.4% | 37.1% |
Reasoning Mistral Small leads
Mistral Large: 15.8 (#310), Mistral Small: 19.8 (#250)
| Benchmark | Mistral Large | Mistral Small |
|---|---|---|
| CritPt | 0% | 0% |
| LiveBench Reasoning | 43.5% | 44.8% |
| LMArena Hard Prompts | 1257 | 1335 |
| DTBench | 65.1% | 70.9% |
| LiveBench Data Analysis | 50.1% | 53.7% |
| LMCA | 16.7% | 20.6% |
| LiveBench | 48.4% | 44% |
| SimpleBench | 22.5% | — |
| Kagi LLM Benchmark | — | 37.8% |
| Epoch Capabilities Index | 128.52 | — |
| ForecastBench | 57.1 | — |
Math Mistral Large leads
Mistral Large: 18.2 (#291), Mistral Small: 16.4 (#293)
| Benchmark | Mistral Large | Mistral Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 5.8% |
| LiveBench Math | 42.5% | 39.9% |
| LMArena Math | 1262 | 1341 |
| MATH Level 5 | 50.3% | 46.8% |
| Omni-MATH | 28.1% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Too close to call
Mistral Large: 30.1 (#230), Mistral Small: 31.0 (#222)
| Benchmark | Mistral Large | Mistral Small |
|---|---|---|
| GPQA Diamond | 51.3% | 47.5% |
| Vectara Hallucination Rate | 4.5% | 5.1% |
| LMArena Expert | 1232 | 1291 |
| MMLU | 80% | 68.7% |
| MMLU-Pro | 59.9% | — |
| Confabulations | 21.4% | — |
| GPQA (HELM) | 43.5% | — |
Multimodal Not comparable
Mistral Large: —, Mistral Small: 33.5 (#96)
| Benchmark | Mistral Large | Mistral Small |
|---|---|---|
| LMArena Vision | — | 1142 |
Multilingual Mistral Small leads
Mistral Large: 40.0 (#219), Mistral Small: 45.5 (#169)
| Benchmark | Mistral Large | Mistral Small |
|---|---|---|
| LMArena Non-English | 1237 | 1315 |
| LMArena Chinese | 1240 | 1340 |
| LMArena French | 1325 | 1337 |
| LMArena German | 1254 | 1340 |
| LMArena Japanese | 1188 | 1275 |
| LMArena Korean | 1202 | 1259 |
| LMArena Russian | 1257 | 1324 |
| LMArena Spanish | 1268 | 1346 |
Instruction Following Mistral Large leads
Mistral Large: 67.9 (#191), Mistral Small: 66.4 (#209)
| Benchmark | Mistral Large | Mistral Small |
|---|---|---|
| LiveBench Instruction Following | 67.9% | 63.7% |
| LMArena Instruction Following | 1249 | 1310 |
| IFEval | 87.7% | — |
Long Context Mistral Small leads
Mistral Large: 38.3 (#199), Mistral Small: 40.4 (#156)
| Benchmark | Mistral Large | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1261 | 1327 |
Writing & Preference Mistral Small leads
Mistral Large: 40.7 (#242), Mistral Small: 52.5 (#171)
| Benchmark | Mistral Large | Mistral Small |
|---|---|---|
| LMArena Text | 1266 | 1338 |
| LMArena Creative Writing | 1243 | 1305 |
| LMArena Multi-Turn | 1260 | 1344 |
| LiveBench Language | 39.4% | 30.5% |
| Short-Story Creative Writing | 69% | — |
| EQ-Bench Creative Writing | 985 | — |
| WildBench | 80.1% | — |
Frequently asked questions
Is Mistral Large better than Mistral Small?
Mistral Small is the stronger model overall, scoring 33.4 to 31.9 on the Noometry Index.
Which is cheaper, Mistral Large or Mistral Small?
Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Mistral Large lists at $2 and $6.
Is Mistral Large or Mistral Small better for coding?
They score almost the same on coding (34.3 vs 34.0); test both on your own repository before choosing.
Which has the bigger context window?
Mistral Small does, with 262K tokens against 131K.
How many benchmarks do Mistral Large and Mistral Small share?
37 benchmarks have published results for both models. Mistral Large has 51 scored results on Noometry and Mistral Small has 39.