Meta’s ‘Mango’ and ‘Avocado’: A New AI Push
7
What is the Viqus Verdict?
We evaluate each news story based on its real impact versus its media hype to offer a clear and objective perspective.
AI Analysis:
While the hype surrounding Meta’s AI ambitions has cooled somewhat, this represents a serious strategic pivot with tangible projects and a committed leadership team, suggesting a more sustainable effort than previous announcements.
Article Summary
Meta is intensifying its efforts in the artificial intelligence race with the development of two significant new models, codenamed ‘Mango’ and ‘Avocado.’ The company’s superintelligence lab, spearheaded by Scale AI co-founder Alexandr Wang, is focused on a text-based model designed to improve coding abilities, alongside a new image and video model capable of understanding visual information and performing reasoning and planning. These projects are intended to address Meta’s perceived lagging behind of competitors like OpenAI and Google. Recent internal restructuring and the departure of key researchers, including chief AI scientist Yann Lecun, have further underscored Meta’s struggle to produce a winning AI product. The company’s AI assistant’s usage primarily relies on the existing user base of its social networks, suggesting these new models will be pivotal for Meta’s future AI strategy.Key Points
- Meta is developing a new image and video model, ‘Mango,’ for release in early 2026.
- A new text-based model, ‘Avocado,’ is also under development, focused on coding capabilities and understanding visual information.
- Meta's strategy relies heavily on these models to address recent internal struggles and regain competitiveness in the AI landscape.