Our brands
Fine-tuning involves re-training an existing model on specific data to adapt it to a precise use.
Fine-tuning starts from an already trained model and specialises it by continuing its training on a targeted dataset. This adapts a general-purpose model to a particular domain or tone.
It is an alternative or a complement to prompting and RAG: rather than providing the context with every request, you inscribe it into the model's weights.
Fine-tuning is relevant when you have many quality examples and are seeking a highly consistent response on a specific case.
Nos experts déploient ces concepts dans des solutions concrètes pour votre entreprise.
