Model comparison
Magistral Small vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.3 to 30.2 on the Noometry Index. Magistral Small costs 4.0× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Last verified . 6 shared benchmarks.
Mistral AISummary
- They share 6 benchmarks with published results for both. Magistral Small scores higher in 2 categories and Mistral Medium in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Mistral Medium leads 24.0 to 6.8.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 6.3% for Magistral Small and 50% for Mistral Medium.
- Magistral Small is cheaper at $0.50 / $1.50 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- Mistral Medium accepts more context: 262K tokens versus 128K.
Side by side
| Magistral Small | Mistral Medium | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 30.2 | 36.3 |
| Released | 2025-06-10 | 2023-12-11 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 40K | 262K |
| Input $ / M tokens | $0.50 | $1.50 |
| Output $ / M tokens | $1.50 | $7.50 |
| Results tracked | 10 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Magistral Small leads
Magistral Small: 38.4 (#176), Mistral Medium: 34.2 (#243)
| Benchmark | Magistral Small | Mistral Medium |
|---|---|---|
| SciCode | 35.2% | 40.2% |
| FrontierCode | — | 8% |
| WeirdML | — | 43.7% |
| LMArena Coding | — | 1434 |
| ALE-Bench | — | 763.98 |
Agentic & Tool Use Not comparable
Magistral Small: —, Mistral Medium: 28.3 (#90)
| Benchmark | Magistral Small | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.7% |
Reasoning Mistral Medium leads
Magistral Small: 6.8 (#350), Mistral Medium: 24.0 (#167)
| Benchmark | Magistral Small | Mistral Medium |
|---|---|---|
| Kagi LLM Benchmark | 6.3% | 50% |
| CritPt | 0.3% | 0% |
| DTBench | 61.3% | 75.5% |
| ARC-AGI-2 | 0% | — |
| ARC-AGI-1 | 5% | — |
| Chess Puzzles | 3% | — |
| LMArena Hard Prompts | — | 1426 |
| LMCA | — | 26.1% |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 133.19 | — |
Math Mistral Medium leads
Magistral Small: 26.2 (#261), Mistral Medium: 28.1 (#245)
| Benchmark | Magistral Small | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 30% | 32.2% |
| ProofBench | — | 9% |
| LMArena Math | — | 1408 |
| MATH Level 5 | — | 81.6% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Magistral Small leads
Magistral Small: 30.9 (#223), Mistral Medium: 25.0 (#265)
| Benchmark | Magistral Small | Mistral Medium |
|---|---|---|
| GPQA Diamond | 56.1% | 59.5% |
| Humanity's Last Exam | — | 4.5% |
| Vectara Hallucination Rate | — | 22.7% |
| LMArena Expert | — | 1408 |
Multimodal Not comparable
Magistral Small: —, Mistral Medium: 35.3 (#88)
| Benchmark | Magistral Small | Mistral Medium |
|---|---|---|
| LMArena Vision | — | 1172 |
Multilingual Not comparable
Magistral Small: —, Mistral Medium: 52.1 (#91)
| Benchmark | Magistral Small | Mistral Medium |
|---|---|---|
| LMArena Non-English | — | 1408 |
| LMArena Chinese | — | 1447 |
| LMArena French | — | 1459 |
| LMArena German | — | 1432 |
| LMArena Japanese | — | 1378 |
| LMArena Korean | — | 1380 |
| LMArena Russian | — | 1411 |
| LMArena Spanish | — | 1433 |
Instruction Following Not comparable
Magistral Small: —, Mistral Medium: 73.7 (#116)
| Benchmark | Magistral Small | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | — | 1398 |
Long Context Not comparable
Magistral Small: —, Mistral Medium: 42.9 (#114)
| Benchmark | Magistral Small | Mistral Medium |
|---|---|---|
| LMArena Longer Query | — | 1406 |
Writing & Preference Not comparable
Magistral Small: —, Mistral Medium: 60.0 (#103)
| Benchmark | Magistral Small | Mistral Medium |
|---|---|---|
| LMArena Text | — | 1424 |
| LMArena Creative Writing | — | 1391 |
| Short-Story Creative Writing | — | 77.3% |
| LMArena Multi-Turn | — | 1418 |
Frequently asked questions
Is Magistral Small better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.3 to 30.2 on the Noometry Index. Magistral Small costs 4.0× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Which is cheaper, Magistral Small or Mistral Medium?
Magistral Small is cheaper. It lists at $0.50 per million input tokens and $1.50 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Magistral Small or Mistral Medium better for coding?
Magistral Small scores higher on coding benchmarks: 38.4 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
Mistral Medium does, with 262K tokens against 128K.
How many benchmarks do Magistral Small and Mistral Medium share?
6 benchmarks have published results for both models. Magistral Small has 10 scored results on Noometry and Mistral Medium has 36.