Model comparison
Magistral Medium vs Mistral Small
Magistral Medium is the stronger model overall, scoring 34.8 to 33.1 on the Noometry Index. Mistral Small costs 10× less per token, which makes it the better buy when Magistral Medium's lead doesn't matter for your workload.
Last verified . 20 shared benchmarks.
Mistral AISummary
- They share 20 benchmarks with published results for both. Magistral Medium scores higher in 3 categories and Mistral Small in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Magistral Medium leads 32.4 to 15.4.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 16.2% for Magistral Medium and 37.8% for Mistral Small.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $2 / $5 for Magistral Medium.
- Mistral Small accepts more context: 256K tokens versus 128K.
Side by side
| Magistral Medium | Mistral Small | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 34.8 | 33.1 |
| Released | 2025-03-17 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 128K | 256K |
| Max output | 16K | 256K |
| Input $ / M tokens | $2 | $0.15 |
| Output $ / M tokens | $5 | $0.60 |
| Results tracked | 22 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Magistral Medium leads
Magistral Medium: 38.6 (#161), Mistral Small: 33.5 (#242)
| Benchmark | Magistral Medium | Mistral Small |
|---|---|---|
| SciCode | 39.2% | 26.5% |
| LMArena Coding | 1319 | 1363 |
| BigCodeBench Instruct | — | 36.1% |
| LiveBench Coding | — | 36.2% |
| BigCodeBench Complete | — | 46.6% |
| ALE-Bench | — | 497.62 |
Agentic & Tool Use Not comparable
Magistral Medium: —, Mistral Small: 27.9 (#87)
| Benchmark | Magistral Medium | Mistral Small |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.1% |
Reasoning Mistral Small leads
Magistral Medium: 8.7 (#342), Mistral Small: 19.3 (#238)
| Benchmark | Magistral Medium | Mistral Small |
|---|---|---|
| Kagi LLM Benchmark | 16.2% | 37.8% |
| CritPt | 0.3% | 0% |
| LMArena Hard Prompts | 1267 | 1336 |
| ARC-AGI-2 | 0% | — |
| ARC-AGI-1 | 6.1% | — |
| LiveBench Reasoning | — | 44.8% |
| DTBench | — | 70.9% |
| LiveBench Data Analysis | — | 53.7% |
| LMCA | — | 20.6% |
| LiveBench | — | 44% |
Math Magistral Medium leads
Magistral Medium: 32.4 (#187), Mistral Small: 15.4 (#287)
| Benchmark | Magistral Medium | Mistral Small |
|---|---|---|
| LMArena Math | 1249 | 1341 |
| OTIS Mock AIME 2024-2025 | — | 5.8% |
| LiveBench Math | — | 39.9% |
| MATH Level 5 | — | 46.8% |
Knowledge Magistral Medium leads
Magistral Medium: 32.9 (#199), Mistral Small: 30.9 (#214)
| Benchmark | Magistral Medium | Mistral Small |
|---|---|---|
| LMArena Expert | 1222 | 1293 |
| GPQA Diamond | — | 47.5% |
| Vectara Hallucination Rate | — | 5.1% |
| MMLU | — | 68.7% |
Multimodal Not comparable
Magistral Medium: —, Mistral Small: 32.9 (#91)
| Benchmark | Magistral Medium | Mistral Small |
|---|---|---|
| LMArena Vision | — | 1142 |
Multilingual Mistral Small leads
Magistral Medium: 39.8 (#219), Mistral Small: 45.7 (#164)
| Benchmark | Magistral Medium | Mistral Small |
|---|---|---|
| LMArena Non-English | 1232 | 1315 |
| LMArena Chinese | 1225 | 1342 |
| LMArena French | 1268 | 1337 |
| LMArena German | 1248 | 1341 |
| LMArena Japanese | 1176 | 1275 |
| LMArena Korean | 1125 | 1260 |
| LMArena Russian | 1224 | 1324 |
| LMArena Spanish | 1269 | 1346 |
Instruction Following Too close to call
Magistral Medium: 66.2 (#207), Mistral Small: 66.6 (#204)
| Benchmark | Magistral Medium | Mistral Small |
|---|---|---|
| LMArena Instruction Following | 1254 | 1310 |
| LiveBench Instruction Following | — | 63.7% |
Long Context Mistral Small leads
Magistral Medium: 39.3 (#178), Mistral Small: 40.4 (#151)
| Benchmark | Magistral Medium | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1295 | 1327 |
Writing & Preference Mistral Small leads
Magistral Medium: 46.5 (#214), Mistral Small: 52.6 (#165)
| Benchmark | Magistral Medium | Mistral Small |
|---|---|---|
| LMArena Text | 1255 | 1338 |
| LMArena Creative Writing | 1245 | 1304 |
| LMArena Multi-Turn | 1275 | 1344 |
| LiveBench Language | — | 30.5% |
Frequently asked questions
Is Magistral Medium better than Mistral Small?
Magistral Medium is the stronger model overall, scoring 34.8 to 33.1 on the Noometry Index. Mistral Small costs 10× less per token, which makes it the better buy when Magistral Medium's lead doesn't matter for your workload.
Which is cheaper, Magistral Medium or Mistral Small?
Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Magistral Medium lists at $2 and $5.
Is Magistral Medium or Mistral Small better for coding?
Magistral Medium scores higher on coding benchmarks: 38.6 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
Mistral Small does, with 256K tokens against 128K.
How many benchmarks do Magistral Medium and Mistral Small share?
20 benchmarks have published results for both models. Magistral Medium has 22 scored results on Noometry and Mistral Small has 39.