Mistral AI models
Ministral all-rounders and the Devstral coding line.
Run on your device
Ministral 3
Mistral AI · Dec 2025Mistral's latest small model family. Fast inference, great for on-device use.
from 2.2 GBruns from 8 GB RAM256K contextApache 2.0
Devstral Small 2
Mistral AI · Dec 2025Mistral's agentic coding model. Tops open models on SWE-bench at its size — purpose-built for software engineering and code agents.
from 14.3 GBruns from 16 GB RAM384K contextApache 2.0