The meta platform opens a new tab on Saturday and released the latest version of its big language model (LLM) Lama, called The Lama 4 Scouts and Lama 4 Mavric.

Meta said that Lama is a multimodal AI system. Multimodal systems are capable of processing and integrating various types of data, including texts, videos, images and audio, and can convert materials into these formats.

Meta said in a statement that the Lama 4 Scouts and Lama 4 Mavric are its “most advanced models” and “best in their classroom for versatile.

Meta said that the Lama 4 MAVIRIC and Lama 4 Scouts will be open source software. It also said that this Lama was previewing 4 beamoths, which is “one of the most clever LLM in the world and our most powerful to serve as a teacher for our new models are our most powerful yet our most powerful.”

Big Technology firms have aggressively invested in Artificial Intelligence (AI) Infrastructure after the success of OpenAI’s chatgipt, which changed the technical scenario and tried to invest in machine learning.
On Friday, it was reported that Meta delayed the launch of the latest version of its LLM as during development, Lama 4 did not meet the expectations of Meta at the Technical Benchmark, especially in logic and mathematics.

The report also concerned that the Lama was less capable than the model of OpenaiI in 4 human voice interactions, stated in the report.

Meta plans to spend this year as $ 65 billion to expand its AI infrastructure, amidst the pressure from investors on Big Tech firms to show returns on its investment.