Meta Launches Llama API, Enhances Developer Access to AI Models

Meta has unveiled several major AI initiatives at its inaugural LlamaCon developer event, including the launch of the Llama API in limited preview and the introduction of a standalone Meta AI app. The Llama API gives developers access to Meta’s Llama models, starting with Llama 3.3 8B, and includes tools for fine-tuning, evaluation, and easy migration from platforms like OpenAI, with assurances that user data submitted via the API will not be used to train Meta's foundation models. The API also enables developers to move their custom models to other providers, and SDKs in Python and Typescript are available to support integration. Additionally, Meta released a free Meta AI app, powered by its Llama 4 model, designed for consumers to interact with the assistant through voice or text, and to receive personalized responses based on their Meta profiles and activity. The app emphasizes user control over shared data and offers features such as multitasking, a Discover feed, and improved voice capabilities. These moves position Meta to compete directly with established AI platforms like OpenAI and Google, while seeking to appeal to both developers and consumers through flexibility and enhanced privacy assurances.