Meta Announces Llama 3 Open-Source AI Model Launch Within a Month
Meta confirms plans to release its next-generation Llama 3 AI model in the coming weeks, featuring multiple versions with enhanced capabilities.
Meta officially confirms that it plans to release Llama 3, its next-generation open-source large language model, within the next month. Speaking at an event in London, Meta’s president of global affairs Nick Clegg states that the company hopes to start rolling out a new suite of foundation models very soon. Rather than a single release, Meta plans to introduce several different iterations of Llama 3 throughout the year, each offering varying capabilities and versatilities.
This upcoming release represents a significant push by Meta to compete directly with OpenAI and integrate advanced AI across its entire product ecosystem. Previous versions of Llama face criticism for being too limited, but Llama 3 addresses these shortcomings by handling a wider range of questions, including more controversial topics, with greater accuracy. The new model reportedly features around 140 billion parameters, which is double the size of the largest Llama 2 model.
By maintaining an open-source approach for the Llama family, Meta deliberately contrasts itself with competitors relying on proprietary models to win favor with developers. Although the company does not provide public demos or discuss specific technical details during the event, executives express high ambitions for the technology. Meta's vice president of AI Research, Joelle Pineau, notes that the ultimate goal is to make a Llama-powered Meta AI the most useful assistant in the world, even while acknowledging that substantial work remains to achieve that milestone.