Mistral AIMagistral Medium
Magistral Medium is a Mistral AI proprietary reasoning model, released 10 Jun 2025. Its best configuration (medium reasoning effort) ranks #617 of 785 on the BenchLeader Index at 41.8 ±5.8, in the lower half. The ± is the point: 331 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in coding (51) and lowest in long context (23). It has been superseded by Magistral Medium 1.2, which is the version to compare against today. Last measured 22 Oct 2026.
- Blended price
- –
- Output speed
- –
- First answer
- –
- Context
- 40k
- Full answer
- –
- Cost per run
- –
- Released
- 10 Jun 2025
- Overall index42
- Reasoning34
- Coding51
- Agents & tools42
- Maths45
- Knowledge49
- Instruction following30
- Human preference46
- Long context23
- Composite40
Versions
Mistral AI has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Magistral Medium 1.2 | 18 Sept 2025 | 43.9 | #558 |
| Magistral Mediummediumthis page | 10 Jun 2025 | 41.8 | #617 |
How its index has moved
8 Sept 2026 to 21 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| Arena Hard Prompts | 1332#228 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 67.9%#318 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 9.8%#328 | Artificial Analysis | |
| Kagi LLM Benchmark | 16.2%#133 | Kagi LLM Benchmark | |
| ARC-AGI-1 | 6.1%#239 | ARC Prize | |
| ARC-AGI-2 | 0.0%#242 | ARC Prize | |
| CritPt | 0.3%#271 | Artificial Analysis |
Coding
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| Arena Coding | 1387#205 | Arena (LMArena) |
Agents & tools
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| Terminal-Bench Hard | 9.1%#231 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 23.1%#307 | Artificial Analysis |
Maths
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| Arena Maths | 1287#239 | Arena (LMArena) |
Knowledge
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| AA-Omniscience | -26.9#268 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 20.3%#300 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 40.7%#157 | Artificial Analysis |
Instruction following
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| IFBench | 25.1%#393 | Artificial Analysis | |
| Arena Instruction Followingnot in index | 1301#235 | Arena (LMArena) |
Human preference
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| Arena Text | 1305#251 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1280#237 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1313#232 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1337#219 | Arena (LMArena) |
Long context
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| AA-LCR | 0.0%#493 | Artificial Analysis |
Composite
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| AA Intelligence Index v4.3.2 | 9.1#411 | Artificial Analysis |
More from Mistral AI
See also
Questions people ask
- What are Magistral Medium's benchmark scores?
- As of 25 Oct 2026, Magistral Medium (medium effort) scores 41.8 ±5.8 on the BenchLeader Index, strongest in coding (51) and weakest in long context (23). That rests on 13 independent benchmark results, and every score on this page links to its publisher.
- How does Magistral Medium rank?
- Magistral Medium (medium effort) is #617 of 785 ranked configurations on 25 Oct 2026. 331 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- When was Magistral Medium released?
- Mistral AI released Magistral Medium on 10 Jun 2025.
Data as of 25 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Magistral Medium: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/magistral-medium.