Mistral Launches Forge for Enterprise AI Model Retraining
Announced at NVIDIA GTC, Forge enables enterprises to retrain frontier-grade AI models on proprietary data — going beyond fine-tuning to fundamental model retraining.
Mistral unveiled Forge at NVIDIA GTC, a platform that lets enterprises and governments retrain frontier-grade AI models on their own proprietary data. This goes significantly beyond typical fine-tuning or RAG approaches — Forge enables fundamental model retraining, giving organizations ownership of customized frontier models.
The platform targets regulated industries like defense, healthcare, and finance where data sovereignty is critical and sending proprietary information to third-party APIs is not an option. Forge works with Mistral's open-weight models as the starting point.
For enterprise AI teams, Forge represents a shift from API-dependent workflows to true model ownership. Organizations can build domain-specific models that retain frontier-level capabilities while being trained exclusively on approved data sources.