Meta's AI Strategy Shift: Llama to Avocado

Meta's AI ambitions are undergoing a notable strategic adjustment, with a perceived shift away from its widely publicized Llama models. Recent indications suggest a renewed focus on a next-generation frontier AI model, internally codenamed Avocado. This strategic redirection is accompanied by substantial investments in hiring top AI talent, aiming to challenge industry leaders such as OpenAI, Google, and Anthropic.

The company's approach to artificial intelligence now appears more concentrated on acquiring specialized expertise rather than solely on its open-source large language models. Insiders reveal that the Avocado model's debut, initially anticipated before the end of 2025, has been rescheduled to the first quarter of 2026. This delay is attributed to ongoing performance testing and training refinements.

As Meta navigates this evolving AI landscape, its significant expenditure on AI talent, including the acquisition of Scale AI's founder and engineers, draws Wall Street's attention. Analysts are closely monitoring the company's progress and its capacity to demonstrate a clear return on these substantial investments amidst a competitive market characterized by rapid adoption of rival AI technologies.