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. | 40k | 1M |
| These models have no shared benchmark scores. | ||
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. | — | 57% |
| Overview | ||
| Company | Mistral | SpaceXSI |
| Release date | Jun 10 2025 | Feb 17 2025 |
| Access | Closed | Closed |
| Model details | View model | View model |
Other comparisons
Magistral MediumvsClaude Haiku 5.5Grok 3vsClaude Haiku 5.5Magistral MediumvsGPT-6.1 TerraGrok 3vsGPT-6.1 TerraMagistral MediumvsGemini 4 CarbonGrok 3vsGemini 4 CarbonMagistral MediumvsMuse Spark OpenGrok 3vsMuse Spark OpenMagistral MediumvsDeepSeek-V4.1-Flash-1017Grok 3vsDeepSeek-V4.1-Flash-1017Magistral MediumvsKimi K3.1 ProGrok 3vsKimi K3.1 ProFrequently asked questions
Magistral Medium and Grok 3 don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Grok 3 shipped 113 days before Magistral Medium, so benchmark comparisons should account for the intervening progress.
Context windows are 40k (Magistral Medium) vs 1M (Grok 3).
Direct benchmark comparisons are unavailable — Magistral Medium and Grok 3 don't publish scores on any of the same benchmarks.