Model comparison
Mistral Large vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.3 to 31.9 on the Noometry Index.
Last verified . 29 shared benchmarks.
Mistral AISummary
- They share 29 benchmarks with published results for both. Mistral Large scores higher in 3 categories and Mistral Medium in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium leads 60.0 to 40.7.
- The biggest single-benchmark swing is MATH Level 5: 50.3% for Mistral Large and 81.6% for Mistral Medium.
- Both cost about the same: $2 input and $6 output per million tokens.
- Mistral Medium accepts more context: 262K tokens versus 131K.
Side by side
| Mistral Large | Mistral Medium | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 31.9 | 36.3 |
| Released | 2024-02-26 | 2023-12-11 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 16K | 262K |
| Input $ / M tokens | $2 | $1.50 |
| Output $ / M tokens | $6 | $7.50 |
| Results tracked | 51 | 36 |
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 Medium: 34.2 (#243)
| Benchmark | Mistral Large | Mistral Medium |
|---|---|---|
| SciCode | 36.2% | 40.2% |
| LMArena Coding | 1277 | 1434 |
| ALE-Bench | 264.7 | 763.98 |
| FrontierCode | — | 8% |
| WeirdML | — | 43.7% |
| BigCodeBench Instruct | 30% | — |
| LiveBench Coding | 47.1% | — |
| BigCodeBench Complete | 38.3% | — |
| HumanEval+ | 62.2% | — |
| MBPP+ | 59.5% | — |
Agentic & Tool Use Too close to call
Mistral Large: 28.6 (#89), Mistral Medium: 28.3 (#90)
| Benchmark | Mistral Large | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | 38.4% | 37.7% |
Reasoning Mistral Medium leads
Mistral Large: 15.8 (#310), Mistral Medium: 24.0 (#167)
| Benchmark | Mistral Large | Mistral Medium |
|---|---|---|
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1257 | 1426 |
| DTBench | 65.1% | 75.5% |
| LMCA | 16.7% | 26.1% |
| SimpleBench | 22.5% | — |
| Kagi LLM Benchmark | — | 50% |
| LiveBench Reasoning | 43.5% | — |
| LiveBench Data Analysis | 50.1% | — |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 128.52 | — |
| ForecastBench | 57.1 | — |
| LiveBench | 48.4% | — |
Math Mistral Medium leads
Mistral Large: 18.2 (#291), Mistral Medium: 28.1 (#245)
| Benchmark | Mistral Large | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 32.2% |
| LMArena Math | 1262 | 1408 |
| MATH Level 5 | 50.3% | 81.6% |
| FrontierMath (Feb 2025 set) | 0.3% | 0.3% |
| ProofBench | — | 9% |
| Omni-MATH | 28.1% | — |
| LiveBench Math | 42.5% | — |
Knowledge Mistral Large leads
Mistral Large: 30.1 (#230), Mistral Medium: 25.0 (#265)
| Benchmark | Mistral Large | Mistral Medium |
|---|---|---|
| GPQA Diamond | 51.3% | 59.5% |
| Vectara Hallucination Rate | 4.5% | 22.7% |
| LMArena Expert | 1232 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| MMLU-Pro | 59.9% | — |
| Confabulations | 21.4% | — |
| GPQA (HELM) | 43.5% | — |
| MMLU | 80% | — |
Multimodal Not comparable
Mistral Large: —, Mistral Medium: 35.3 (#88)
| Benchmark | Mistral Large | Mistral Medium |
|---|---|---|
| LMArena Vision | — | 1172 |
Multilingual Mistral Medium leads
Mistral Large: 40.0 (#219), Mistral Medium: 52.1 (#91)
| Benchmark | Mistral Large | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1237 | 1408 |
| LMArena Chinese | 1240 | 1447 |
| LMArena French | 1325 | 1459 |
| LMArena German | 1254 | 1432 |
| LMArena Japanese | 1188 | 1378 |
| LMArena Korean | 1202 | 1380 |
| LMArena Russian | 1257 | 1411 |
| LMArena Spanish | 1268 | 1433 |
Instruction Following Mistral Medium leads
Mistral Large: 67.9 (#191), Mistral Medium: 73.7 (#116)
| Benchmark | Mistral Large | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1249 | 1398 |
| LiveBench Instruction Following | 67.9% | — |
| IFEval | 87.7% | — |
Long Context Mistral Medium leads
Mistral Large: 38.3 (#199), Mistral Medium: 42.9 (#114)
| Benchmark | Mistral Large | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1261 | 1406 |
Writing & Preference Mistral Medium leads
Mistral Large: 40.7 (#242), Mistral Medium: 60.0 (#103)
| Benchmark | Mistral Large | Mistral Medium |
|---|---|---|
| LMArena Text | 1266 | 1424 |
| LMArena Creative Writing | 1243 | 1391 |
| Short-Story Creative Writing | 69% | 77.3% |
| LMArena Multi-Turn | 1260 | 1418 |
| EQ-Bench Creative Writing | 985 | — |
| WildBench | 80.1% | — |
| LiveBench Language | 39.4% | — |
Frequently asked questions
Is Mistral Large better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.3 to 31.9 on the Noometry Index.
Which is cheaper, Mistral Large or Mistral Medium?
Mistral Medium is cheaper. It lists at $1.50 per million input tokens and $7.50 per million output tokens; Mistral Large lists at $2 and $6.
Is Mistral Large or Mistral Medium better for coding?
They score almost the same on coding (34.3 vs 34.2); test both on your own repository before choosing.
Which has the bigger context window?
Mistral Medium does, with 262K tokens against 131K.
How many benchmarks do Mistral Large and Mistral Medium share?
29 benchmarks have published results for both models. Mistral Large has 51 scored results on Noometry and Mistral Medium has 36.