Model comparison
Magistral Medium vs Mistral Large
Magistral Medium is the stronger model overall, scoring 35.2 to 31.9 on the Noometry Index.
Last verified . 19 shared benchmarks.
Mistral AISummary
- They share 19 benchmarks with published results for both. Magistral Medium scores higher in 5 categories and Mistral Large in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Magistral Medium leads 35.1 to 18.2.
- Magistral Medium is cheaper at $2 / $5 per million input/output tokens, against $2 / $6 for Mistral Large.
- Magistral Medium accepts more context: 262K tokens versus 131K.
Side by side
| Magistral Medium | Mistral Large | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 35.2 | 31.9 |
| Released | 2025-03-17 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 16K | 16K |
| Input $ / M tokens | $2 | $2 |
| Output $ / M tokens | $5 | $6 |
| Results tracked | 22 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Magistral Medium leads
Magistral Medium: 39.1 (#161), Mistral Large: 34.3 (#240)
| Benchmark | Magistral Medium | Mistral Large |
|---|---|---|
| SciCode | 39.2% | 36.2% |
| LMArena Coding | 1319 | 1277 |
| BigCodeBench Instruct | — | 30% |
| LiveBench Coding | — | 47.1% |
| BigCodeBench Complete | — | 38.3% |
| ALE-Bench | — | 264.7 |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use Not comparable
Magistral Medium: —, Mistral Large: 28.6 (#89)
| Benchmark | Magistral Medium | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 38.4% |
Reasoning Mistral Large leads
Magistral Medium: 8.6 (#348), Mistral Large: 15.8 (#310)
| Benchmark | Magistral Medium | Mistral Large |
|---|---|---|
| CritPt | 0.3% | 0% |
| LMArena Hard Prompts | 1267 | 1257 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | — | 22.5% |
| Kagi LLM Benchmark | 16.2% | — |
| ARC-AGI-1 | 6.1% | — |
| LiveBench Reasoning | — | 43.5% |
| DTBench | — | 65.1% |
| LiveBench Data Analysis | — | 50.1% |
| LMCA | — | 16.7% |
| Epoch Capabilities Index | — | 128.52 |
| ForecastBench | — | 57.1 |
| LiveBench | — | 48.4% |
Math Magistral Medium leads
Magistral Medium: 35.1 (#189), Mistral Large: 18.2 (#291)
| Benchmark | Magistral Medium | Mistral Large |
|---|---|---|
| LMArena Math | 1250 | 1262 |
| OTIS Mock AIME 2024-2025 | — | 8.5% |
| Omni-MATH | — | 28.1% |
| LiveBench Math | — | 42.5% |
| MATH Level 5 | — | 50.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Magistral Medium leads
Magistral Medium: 33.5 (#202), Mistral Large: 30.1 (#230)
| Benchmark | Magistral Medium | Mistral Large |
|---|---|---|
| LMArena Expert | 1223 | 1232 |
| GPQA Diamond | — | 51.3% |
| MMLU-Pro | — | 59.9% |
| Confabulations | — | 21.4% |
| Vectara Hallucination Rate | — | 4.5% |
| GPQA (HELM) | — | 43.5% |
| MMLU | — | 80% |
Multilingual Too close to call
Magistral Medium: 39.6 (#224), Mistral Large: 40.0 (#219)
| Benchmark | Magistral Medium | Mistral Large |
|---|---|---|
| LMArena Non-English | 1232 | 1237 |
| LMArena Chinese | 1227 | 1240 |
| LMArena French | 1267 | 1325 |
| LMArena German | 1248 | 1254 |
| LMArena Japanese | 1175 | 1188 |
| LMArena Korean | 1125 | 1202 |
| LMArena Russian | 1224 | 1257 |
| LMArena Spanish | 1271 | 1268 |
Instruction Following Mistral Large leads
Magistral Medium: 66.0 (#211), Mistral Large: 67.9 (#191)
| Benchmark | Magistral Medium | Mistral Large |
|---|---|---|
| LMArena Instruction Following | 1254 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
| IFEval | — | 87.7% |
Long Context Magistral Medium leads
Magistral Medium: 39.3 (#183), Mistral Large: 38.3 (#199)
| Benchmark | Magistral Medium | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1295 | 1261 |
Writing & Preference Magistral Medium leads
Magistral Medium: 46.3 (#219), Mistral Large: 40.7 (#242)
| Benchmark | Magistral Medium | Mistral Large |
|---|---|---|
| LMArena Text | 1255 | 1266 |
| LMArena Creative Writing | 1245 | 1243 |
| LMArena Multi-Turn | 1275 | 1260 |
| Short-Story Creative Writing | — | 69% |
| EQ-Bench Creative Writing | — | 985 |
| WildBench | — | 80.1% |
| LiveBench Language | — | 39.4% |
Frequently asked questions
Is Magistral Medium better than Mistral Large?
Magistral Medium is the stronger model overall, scoring 35.2 to 31.9 on the Noometry Index.
Which is cheaper, Magistral Medium or Mistral Large?
Magistral Medium is cheaper. It lists at $2 per million input tokens and $5 per million output tokens; Mistral Large lists at $2 and $6.
Is Magistral Medium or Mistral Large better for coding?
Magistral Medium scores higher on coding benchmarks: 39.1 versus 34.3 in the Noometry coding category.
Which has the bigger context window?
Magistral Medium does, with 262K tokens against 131K.
How many benchmarks do Magistral Medium and Mistral Large share?
19 benchmarks have published results for both models. Magistral Medium has 22 scored results on Noometry and Mistral Large has 51.