Meta is gearing up for a major AI expansion, with two new flagship artificial intelligence models reportedly on track to arrive in the first half of 2026. The move signals the company’s most ambitious bid yet to strengthen its position in the fast-moving generative AI race and narrow the gap with leading competitors.
According to a report from The Wall Street Journal, Meta is developing an image and video generation model internally nicknamed “Mango.” This system is expected to focus on creating and transforming visual content, an increasingly high-demand area as creators, brands, and everyday users look for faster ways to produce short-form videos, promotional visuals, and social-ready media using AI.
Alongside Mango, Meta is also said to be building a text-focused large language model (LLM) called “Avocado.” While details remain limited, a next-generation text LLM typically points to improvements in conversational accuracy, reasoning, writing quality, summarization, and useful task completion—capabilities that power AI assistants across apps, search, and productivity tools.
If both models launch on schedule in early 2026, it would represent a significant step in Meta’s broader AI strategy. Image generation, video creation, and advanced text intelligence are now central pillars of modern AI platforms, and developing dedicated flagship models in each category suggests Meta is aiming to compete more directly at the highest tier of generative AI.
With Mango and Avocado, Meta appears to be positioning itself for a sharper, more competitive push into AI-driven creation and communication—setting up 2026 as an important year for the company’s next wave of AI products and experiences.
