Aiming to overtake OpenAI, the leader in the generative AI industry, Meta Platforms unveiled early versions of its most recent large language model, Llama 3, on Thursday. It also released an image generator that updates visuals in real time while users input prompts.
The company presents Meta AI, a free virtual assistant, as the most advanced among its competitors. To support this claim, it has included performance comparisons between Meta AI and rival products from Alphabet’s Google and the French startup Mistral AI, on topics like reasoning, coding, and creative writing.
As part of its drive to catch up, the social media behemoth has started freely giving its Llama models to developers creating AI apps. This potent free option might thwart competitors’ plans to make money off their proprietary technology. Critics concerned about what unethical individuals might develop with the concept have expressed worries about safety related to the method.
In an interview, Meta Chief Product Officer Chris Cox stated that the company has given Llama 3 new computer coding skills and that it was fed both text and images during training. Nevertheless, for the time being, the model produces only text.
He said that future iterations will include more sophisticated reasoning, such as the capacity to create longer multi-step plans. Release versions scheduled over the upcoming months.
Also Read:
Emirates Suspends Check-in for Travelers who have Stopovers in Dubai.
Infosys Q4 2024 Results: IT Behemoth Declares Special Dividend
