Android

Meta announces LlamaCon, its first dev conference focused on GenAI


Meta today announced LlamaCon, which is the company’s first-ever generative AI developer conference. The conference picks its name from Meta’s Llama family of generative AI models. At the conference, Meta will focus on sharing the latest on its open-source AI development to help developers create amazing apps and products.

Meta announces the annual Connect event and LlamaCon developer conference

Meta will hold LlamaCon on April 29, 2025. The company has also noted that it will share more details about the conference in the coming days. In addition to the conference, Meta announced dates for its annual Connect event. Meta Connect will take place later in the year, on September 17 and 18.

The announcement of LlamaCon seems feasible given that Meta has long been a believer in open-source AI. However, the company is reportedly worried about the sudden rise of DeepSeek’s AI model, which could reportedly beat the next version of Meta’s Llama AI. Late last month, a paywalled report by The Information mentioned that Meta started working with a dedicated team to find out why DeepSeek’s AI is so good.

Well, it’s true that Meta is trying to compete aggressively in the ongoing AI race. That’s one of the main reasons why it announced plans to invest up to $80 billion in numerous AI-related projects. The investment, announced before LlamaCon, will help Meta build AI data centers and hire new talents from the AI industry.

Meta’s new Llama AI models could launch in the coming months

Meta CEO Mark Zuckerberg previously said that the company could launch several Llama models in the upcoming months. Those might include reasoning models capable of competing with OpenAI’s o3-mini and other models having native multimodal capabilities. Meta may take the center stage of LlamaCon to unveil its new AI models—who knows?

While Meta is trying to cement its place in the AI space, it has repeatedly gotten into legal troubles. More recently, a lawsuit accuses Meta of torrenting 82TB of copyrighted books to train its Llama AI models. Previously, many countries in the EU have forced Meta to withhold its next multimodal AI model over data privacy concerns.



READ SOURCE

This website uses cookies. By continuing to use this site, you accept our use of cookies.