Model comparison
Magistral Medium vs o3
o3 is the stronger model overall, scoring 47.5 to 35.2 on the Noometry Index.
Last verified . 21 shared benchmarks.
Mistral AISummary
- They share 21 benchmarks with published results for both. Magistral Medium scores higher in 0 categories and o3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where o3 leads 32.0 to 8.6.
- The biggest single-benchmark swing is ARC-AGI-1: 6.1% for Magistral Medium and 60.8% for o3.
- Magistral Medium is cheaper at $2 / $5 per million input/output tokens, against $2 / $8 for o3.
- Magistral Medium accepts more context: 262K tokens versus 200K.
- Magistral Medium has downloadable open weights; the other is API-only.
Side by side
| Magistral Medium | o3 | |
|---|---|---|
| Provider | Mistral AI | OpenAI |
| Noometry Index | 35.2 | 47.5 |
| Released | 2025-03-17 | 2025-04-16 |
| Weights | Open | Proprietary |
| Context window | 262K | 200K |
| Max output | 16K | 100K |
| Input $ / M tokens | $2 | $2 |
| Output $ / M tokens | $5 | $8 |
| Results tracked | 22 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
Magistral Medium: 39.1 (#161), o3: 46.8 (#64)
| Benchmark | Magistral Medium | o3 |
|---|---|---|
| LMArena Coding | 1319 | 1408 |
| SWE-bench Verified | — | 62.3% |
| SWE-bench Verified (bash only) | — | 58.4% |
| Aider Polyglot | — | 81.3% |
| SciCode | 39.2% | — |
| GSO | — | 8.8% |
| WeirdML | — | 52.4% |
| CadEval | — | 74% |
| ALE-Bench | — | 933.55 |
Agentic & Tool Use Not comparable
Magistral Medium: —, o3: 34.5 (#44)
| Benchmark | Magistral Medium | o3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 30.8% |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| LMArena Search | — | 1144 |
| METR Time Horizons | — | 65.4% |
Reasoning o3 leads
Magistral Medium: 8.6 (#348), o3: 32.0 (#78)
| Benchmark | Magistral Medium | o3 |
|---|---|---|
| ARC-AGI-2 | 0% | 6.5% |
| Kagi LLM Benchmark | 16.2% | 67.6% |
| ARC-AGI-1 | 6.1% | 60.8% |
| CritPt | 0.3% | 1.4% |
| LMArena Hard Prompts | 1267 | 1402 |
| SimpleBench | — | 53.1% |
| Chess Puzzles | — | 38% |
| EnigmaEval | — | 13.1% |
| Mystery Game Puzzles | — | 29% |
| DTBench | — | 84.8% |
| LMCA | — | 39.7% |
| Epoch Capabilities Index | — | 146.86 |
| ForecastBench | — | 62.5 |
Math o3 leads
Magistral Medium: 35.1 (#189), o3: 50.2 (#58)
| Benchmark | Magistral Medium | o3 |
|---|---|---|
| LMArena Math | 1250 | 1426 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| OTIS Mock AIME 2024-2025 | — | 84.4% |
| Omni-MATH | — | 71.4% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 18.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge o3 leads
Magistral Medium: 33.5 (#202), o3: 54.6 (#52)
| Benchmark | Magistral Medium | o3 |
|---|---|---|
| LMArena Expert | 1223 | 1402 |
| GPQA Diamond | — | 81.8% |
| Humanity's Last Exam | — | 20.3% |
| SimpleQA Verified | — | 49.4% |
| MMLU-Pro | — | 85.9% |
| Confabulations | — | 14.4% |
| GPQA (HELM) | — | 75.3% |
Multimodal Not comparable
Magistral Medium: —, o3: 41.4 (#36)
| Benchmark | Magistral Medium | o3 |
|---|---|---|
| LMArena Vision | — | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
Multilingual o3 leads
Magistral Medium: 39.6 (#224), o3: 51.7 (#105)
| Benchmark | Magistral Medium | o3 |
|---|---|---|
| LMArena Non-English | 1232 | 1401 |
| LMArena Chinese | 1227 | 1437 |
| LMArena French | 1267 | 1430 |
| LMArena German | 1248 | 1420 |
| LMArena Japanese | 1175 | 1403 |
| LMArena Korean | 1125 | 1370 |
| LMArena Russian | 1224 | 1406 |
| LMArena Spanish | 1271 | 1395 |
Instruction Following o3 leads
Magistral Medium: 66.0 (#211), o3: 72.8 (#127)
| Benchmark | Magistral Medium | o3 |
|---|---|---|
| LMArena Instruction Following | 1254 | 1368 |
| IFEval | — | 86.9% |
Long Context o3 leads
Magistral Medium: 39.3 (#183), o3: 53.3 (#6)
| Benchmark | Magistral Medium | o3 |
|---|---|---|
| LMArena Longer Query | 1295 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |
Writing & Preference o3 leads
Magistral Medium: 46.3 (#219), o3: 63.5 (#64)
| Benchmark | Magistral Medium | o3 |
|---|---|---|
| LMArena Text | 1255 | 1410 |
| LMArena Creative Writing | 1245 | 1359 |
| LMArena Multi-Turn | 1275 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
| EQ-Bench Creative Writing | — | 1676 |
| WildBench | — | 86.1% |
Frequently asked questions
Is Magistral Medium better than o3?
o3 is the stronger model overall, scoring 47.5 to 35.2 on the Noometry Index.
Which is cheaper, Magistral Medium or o3?
Magistral Medium is cheaper. It lists at $2 per million input tokens and $5 per million output tokens; o3 lists at $2 and $8.
Is Magistral Medium or o3 better for coding?
o3 scores higher on coding benchmarks: 46.8 versus 39.1 in the Noometry coding category.
Which has the bigger context window?
Magistral Medium does, with 262K tokens against 200K.
How many benchmarks do Magistral Medium and o3 share?
21 benchmarks have published results for both models. Magistral Medium has 22 scored results on Noometry and o3 has 63.