Meta is reportedly working on new multimedia model with H1 2026 release date in mind

Core Insights - Meta is developing new AI models under its superintelligence lab, including an image and video model codenamed "Mango" and a text-based model known as "Avocado" [1][2] - The company plans to release these models in the first half of 2026, aiming to enhance coding capabilities and create models that can understand visual information and reason without extensive training [2] - Meta has fallen behind competitors like OpenAI, Anthropic, and Google in the AI sector, experiencing significant restructuring and leadership changes within its AI division [3] Company Developments - The internal Q&A revealed by Meta's leadership indicates a strategic roadmap for AI development, focusing on improving the text-based model's coding abilities and exploring new world models [2] - Despite the restructuring efforts, several researchers who joined Meta SuperIntelligence Labs have already left, including the chief AI scientist, Yann Lecun, who is starting his own venture [3] Market Position - Currently, Meta lacks a standout AI product, relying on its vast user base across social networks to support its AI assistant, which is integrated into the search bar of its applications [4] - The success of the upcoming projects and models from Meta SuperIntelligence Labs is critical for the company's future in the AI landscape [4]