Meta Platforms is preparing to release the latest version of its large language model, Llama 4,
later this month, according to a report by The Information. The release has already been delayed at least twice, as the company races to keep pace in the increasingly competitive artificial intelligence space. However, sources familiar with the matter say the launch could be pushed back once again.
One of the main reasons for the delay is that Llama 4 has reportedly fallen short of Meta's technical expectations particularly in areas like reasoning and mathematics. Additionally, the model is said to be less proficient than OpenAI's offerings when it comes to conducting humanlike voice conversations, further contributing to Meta's hesitation.
Despite these challenges, Meta remains committed to its AI ambitions. The company plans to invest up to $65 billion this year to expand its AI infrastructure, responding to mounting investor pressure to demonstrate a clear return on such heavy spending.
Meanwhile, competition from Chinese AI company DeepSeek, which has introduced a strong lower-cost model is challenging the assumption that creating top-tier AI requires billions in investment. Llama 4 is expected to incorporate some technical strategies used by DeepSeek, including the mixture of experts technique, which allows different parts of the model to specialize in specific tasks, enhancing overall performance.
Meta is also reportedly considering a staggered rollout of Llama 4: first making it available through its Meta AI assistant, then potentially releasing it as open-source software at a later date. The company's previous model, Llama 3, launched last year, offered strong capabilities in multiple languages, coding, and complex problem-solving, and was made available mostly for free—highlighting Meta's continued push to make its AI tools widely accessible.
Found this article interesting? Follow us on X(Twitter) and FaceBook to read more exclusive content we post.
