Meta is working on a new AI system focused on image and video creation, internally known as Mango.
At the same time, the company is developing its next major text-based language model. Meta’s chief AI officer, Alexandr Wang, spoke about these projects during an internal question-and-answer session with chief product officer Chris Cox, according to people familiar with the discussion. Both models are expected to be released in the first half of 2026.
Wang explained that the new text model, code-named Avocado, is being built with a strong focus on improving coding abilities. He also said Meta has begun early research into what are known as world models, which are AI systems that learn about their surroundings by processing visual information from the environment.
Earlier this year, Meta reorganized its AI efforts and created a new group called Meta Superintelligence Labs, placing Wang in charge. CEO Mark Zuckerberg personally led a major hiring push, bringing in more than 20 researchers from OpenAI and forming a new team of over 50 people with deep experience in artificial intelligence.
The race to lead in image and video generation has intensified among major tech companies. In September, Meta launched an AI video tool called Vibes, developed in partnership with Midjourney. Within days, OpenAI responded by releasing its own AI video generator, known as Sora.
Other rivals have also seen rapid growth driven by visual AI tools. Google’s image creation feature, Nano Banana, helped increase Gemini’s monthly user base from around 450 million in July to more than 650 million by late October. After Google released a new version of Gemini in November, OpenAI CEO Sam Altman reportedly urged his teams to move faster to regain an edge in key performance areas, leading to the release of an updated ChatGPT image generation tool.
If this article helped you, please consider supporting our work. Every small contribution keeps Abijita.com independent and running.
Altman later told journalists that image generation plays a major role in attracting and retaining users. He said many people are drawn to AI for its visual features, and that these tools are among the strongest reasons users keep coming back. (via WSJ)
