Nvidia has released a new AI model, Llama-3.1-Nemotron-70B-Instruct, that is making waves in the AI community. This smaller model, built on Meta’s Llama 3.1 framework, has surprisingly outperformed larger competitors in benchmark tests. Its efficiency and open-source availability position Nvidia as a major player in AI software, highlighting its shift from hardware to AI development.
NVIDIA has launched AI Foundry, a service that helps companies build custom AI ‘Supermodels’ based on Meta’s open-source Llama 3.1 generative AI model. These models are trained for specific industry needs, leveraging NVIDIA’s software and expertise, and powered by the NVIDIA DGX Cloud AI platform. This partnership signifies a significant step towards widespread adoption of generative AI, empowering businesses to create tailored AI applications.
Meta has announced a significant update to its AI, bringing a smart voice assistant to Meta Quest 3, Quest Pro, and Quest 2 headsets. The update, rolling out in August for US and Canadian users, will enable users to interact with the AI using their voice and even have it see real-world objects through the headset’s camera. This update follows the release of Meta’s powerful Llama 3.1 AI model, which is open source and competitive with other leading AI models.
Facebook’s parent company, Meta, has released Llama 3.1, a new open-source large language model available in three sizes. This release marks a significant step forward in AI accessibility and marks a major advancement for the open-source AI community, with Meta predicting that future models will become industry leaders.
