Meta Platforms(META.O) has unveiled the version of its AI experiment, Llama 3, and the newest real-time image generator ‘IMAGINE’. AI development has always been driving the interest of companies like Meta as they look to match OpenAI – the leader of the Generative AI market. The new models are advanced technologies of The Meta AI suite, which is declared as the first-class and free virtual assistant service, different from other competitors like Google and Mistral.
Meta AI Llama 3 developments across its platforms
The updated version of Meta AI assistant will reach a larger audience due to the company’s integration into popular platforms, namely Facebook, Instagram, WhatsApp, and Messenger. Meta has also launched a website, which is a standalone feature designed to provide users with interactive experiences like creating a packing list for a trip, doing a music quiz, or creating images of city skylines. In this case, Meta will compete against OpenAI from Microsoft, using limited resources to combat.
The change in Meta also implies a substantial change (or upgrade) of Meta’s computing infrastructure and the bringing together Meta’s research and product units. These adaptations are about to make the AI capabilities of the company more accessible yet efficient to its numerous users.
The company has taken a further step forward by releasing its Llama models to developers, providing a robust ML tool with the possibility to put under question the revenue models of its rivals. This transparency, however, has led to some reservations about the technology coming from potentially abusing the technology.
Meta AI development and strategic partnerships toward global expansion
Meta equips Llama 3 with superior coding capability in addition to the new pictorial data in training, but now it produces text instead of images. Meta’s current and upcoming product developments in artificial intelligence are indeed very impressive, according to Meta’s Chief Product Officer, Chris Cox. He emphasized that high-level reasoning skills and dual-output generation of text and image will be included in future releases. The providers expect these new features to be gradually deployed from now on.
Other than that, Meta made a deal with Google to provide up-to-the-minute search results that assistants can use to answer inquiries. This also expanded a collaboration with Microsoft Bing, which is now available. According to Meta’s global AI agenda, this extension of services covers more than a dozen countries outside the US, including Australia, Canada, Singapore, Nigeria, Pakistan, and Europe, where plans are still being worked on due to stringent privacy laws.
Meta market position rises with new AI model
The price of MetaShares bounced back in the afternoon after the statement, climbing by 1.8%. Llama 3’s 8 billion and 70 billion parameter models launched, the additional versions that already beat the performance of the famous free libraries located in the market. Roughly 400 billion weights are being trained in the largest model, which has many other settings to improve AI intelligence.
Source: Reuters
Meta’s decision to offer extra AI tools is a strategic decision that aims to position the tech company to compete in an industry where creativity and fast delivery are crucial.
The constantly changing and mercurial AI landscape allows Meta to boost its investments in advanced capabilities and strategic partners. However, with such a development, Meta could stand a better chance of getting a good field position.
News sourced from meta blogpost
