Model comparison
Codestral vs Magistral Medium
Magistral Medium is the stronger model overall, scoring 35.2 to 30.6 on the Noometry Index. Codestral costs 6.1× less per token, which makes it the better buy when Magistral Medium's lead doesn't matter for your workload.
Last verified . 1 shared benchmarks.
Mistral AISummary
- They share 1 benchmark with published results for both. Codestral scores higher in 1 category and Magistral Medium in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where Magistral Medium leads 39.1 to 27.3.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 32.5% for Codestral and 16.2% for Magistral Medium.
- Codestral is cheaper at $0.30 / $0.90 per million input/output tokens, against $2 / $5 for Magistral Medium.
- Magistral Medium accepts more context: 262K tokens versus 256K.
- Magistral Medium has downloadable open weights; the other is API-only.
Side by side
| Codestral | Magistral Medium | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 30.6 | 35.2 |
| Released | 2024-05-29 | 2025-03-17 |
| Weights | Proprietary | Open |
| Context window | 256K | 262K |
| Max output | 8K | 16K |
| Input $ / M tokens | $0.30 | $2 |
| Output $ / M tokens | $0.90 | $5 |
| Results tracked | 7 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Magistral Medium leads
Codestral: 27.3 (#321), Magistral Medium: 39.1 (#161)
| Benchmark | Codestral | Magistral Medium |
|---|---|---|
| Aider Polyglot | 11.1% | — |
| SciCode | — | 39.2% |
| BigCodeBench Instruct | 41.8% | — |
| LMArena Coding | — | 1319 |
| BigCodeBench Complete | 52.5% | — |
| ALE-Bench | 137.78 | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Reasoning Codestral leads
Codestral: 19.8 (#251), Magistral Medium: 8.6 (#348)
| Benchmark | Codestral | Magistral Medium |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | 16.2% |
| ARC-AGI-2 | — | 0% |
| ARC-AGI-1 | — | 6.1% |
| CritPt | — | 0.3% |
| LMArena Hard Prompts | — | 1267 |
Math Not comparable
Codestral: —, Magistral Medium: 35.1 (#189)
| Benchmark | Codestral | Magistral Medium |
|---|---|---|
| LMArena Math | — | 1250 |
Knowledge Not comparable
Codestral: —, Magistral Medium: 33.5 (#202)
| Benchmark | Codestral | Magistral Medium |
|---|---|---|
| LMArena Expert | — | 1223 |
Multilingual Not comparable
Codestral: —, Magistral Medium: 39.6 (#224)
| Benchmark | Codestral | Magistral Medium |
|---|---|---|
| LMArena Non-English | — | 1232 |
| LMArena Chinese | — | 1227 |
| LMArena French | — | 1267 |
| LMArena German | — | 1248 |
| LMArena Japanese | — | 1175 |
| LMArena Korean | — | 1125 |
| LMArena Russian | — | 1224 |
| LMArena Spanish | — | 1271 |
Instruction Following Not comparable
Codestral: —, Magistral Medium: 66.0 (#211)
| Benchmark | Codestral | Magistral Medium |
|---|---|---|
| LMArena Instruction Following | — | 1254 |
Long Context Not comparable
Codestral: —, Magistral Medium: 39.3 (#183)
| Benchmark | Codestral | Magistral Medium |
|---|---|---|
| LMArena Longer Query | — | 1295 |
Writing & Preference Not comparable
Codestral: —, Magistral Medium: 46.3 (#219)
| Benchmark | Codestral | Magistral Medium |
|---|---|---|
| LMArena Text | — | 1255 |
| LMArena Creative Writing | — | 1245 |
| LMArena Multi-Turn | — | 1275 |
Frequently asked questions
Is Codestral better than Magistral Medium?
Magistral Medium is the stronger model overall, scoring 35.2 to 30.6 on the Noometry Index. Codestral costs 6.1× less per token, which makes it the better buy when Magistral Medium's lead doesn't matter for your workload.
Which is cheaper, Codestral or Magistral Medium?
Codestral is cheaper. It lists at $0.30 per million input tokens and $0.90 per million output tokens; Magistral Medium lists at $2 and $5.
Is Codestral or Magistral Medium better for coding?
Magistral Medium scores higher on coding benchmarks: 39.1 versus 27.3 in the Noometry coding category.
Which has the bigger context window?
Magistral Medium does, with 262K tokens against 256K.
How many benchmarks do Codestral and Magistral Medium share?
1 benchmark has published results for both models. Codestral has 7 scored results on Noometry and Magistral Medium has 22.