Meta Partially Unveils Its AI Models, A Hybrid Strategy
Le brief IA que les pros lisent chaque soir
Les 7 actus IA du jour, décryptées en 5 min. Gratuit.
Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.
Choisis ton rythme
Gratuit · Pas de spam · Désabonnement en 1 clic
Meta, the social media giant, is set to take a new step in the field of artificial intelligence by releasing parts of its new models as open source. According to a report from Axios, this initiative marks a turning point under the leadership of Alexandr Wang, who joined Meta in 2025 through a nearly $15 billion deal with Scale AI.
However, this openness will not be complete. Unlike the approach taken with the Llama models, Meta plans to keep certain components of its new models proprietary. This decision is driven by a rigorous assessment of security risks prior to any release. Furthermore, the most advanced models will not be made available to the general public.
Alexandr Wang sees Meta as a counterbalance to companies like Anthropic and OpenAI, which primarily focus on government and enterprise clients. In contrast, Meta's strategy aims to reach consumers directly through its flagship platforms such as WhatsApp, Facebook, and Instagram. Axios sources emphasize that Meta is aware that its new models will not be competitive in all areas compared to its rivals.
Brief IA — L'actualité IA en français
L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.