magistral-medium-2509
Our frontier-class reasoning model release candidate September 2025.
Input
Output
Context
131K
Max Output
40K
Parameters
—
Technical Specifications
Model TypeVision
Context Window131K
Max Output Tokens40K
Parameters—
Training Cutoff—
Licenseapache-2.0
Capabilities
Input Modalities
textimage
Output Modalities
text
Features
function_callingjson_modereasoningstreamingtext_generationvision
Related Tools
Data sourced from official provider APIs and documentation
Last updated: May 5, 2026
Start building with the right model.
From model selection to production, one platform, no fragmentation.