Compare AI models
| Specifications | ||
Context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider. | 128k | 1M |
| Benchmarks | ||
LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. | 75% | 57% |
AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools | 91.82% | — |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 83.48% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 76.26% | — |
| Overview | ||
| Company | Mistral | SpaceXSI |
| Release date | Sep 17 2025 | Feb 17 2025 |
| Access | Closed | Closed |
| Model details | View model | View model |
Other comparisons
Magistral Medium 1.2vsClaude Fable 5.2Grok 3vsClaude Fable 5.2Magistral Medium 1.2vsGPT-6.2 AstraGrok 3vsGPT-6.2 AstraMagistral Medium 1.2vsGemini 4 HydrogenGrok 3vsGemini 4 HydrogenMagistral Medium 1.2vsMuse EmberGrok 3vsMuse EmberMagistral Medium 1.2vsDeepSeek-V4.1-Flash-1017Grok 3vsDeepSeek-V4.1-Flash-1017Magistral Medium 1.2vsKimi K3.1 ProGrok 3vsKimi K3.1 ProFrequently asked questions
Magistral Medium 1.2 leads Grok 3 on 1 of the 1 benchmark they both report (LiveCodeBench). Grok 3 shipped 212 days before Magistral Medium 1.2, so benchmark comparisons should account for the intervening progress.
Context windows are 128k (Magistral Medium 1.2) vs 1M (Grok 3).
On LiveCodeBench, Magistral Medium 1.2 leads at 75% vs Grok 3 at 57%.