Meta Delays Launch of AI Model "Avocado" to May 2026

META
March 13, 2026

Meta Platforms postponed the launch of its next‑generation AI model, code‑named Avocado, to at least May 2026 after internal testing revealed performance gaps relative to competitors such as Google’s Gemini 3.0 and Anthropic’s Claude in reasoning, coding, and writing tasks.

The delay follows a broader pattern of setbacks in Meta’s AI roadmap. Avocado was intended to serve as a reset for the company’s AI strategy, but the testing results indicated that the model was not yet competitive enough to support the company’s ambitious AI monetization plans, which rely on enhancing ad targeting, recommendation engines, and other platform services.

Meta’s leadership acknowledged the setback, noting that the company is exploring interim solutions, including the possibility of licensing Google’s Gemini technology, to bridge the performance gap while the internal team continues to refine Avocado. The spokesperson emphasized that the delay does not alter the company’s long‑term commitment to building proprietary AI capabilities.

The announcement comes amid a significant capital‑spending push, with Meta earmarking $115 billion to $135 billion for AI infrastructure in 2026, including custom chips and data‑center expansion. Investors are scrutinizing the return on this investment, especially as Meta’s AI model development has faced multiple delays, such as the earlier Llama 4 rollout.

Market reaction to the news was muted, with Meta’s shares falling about 1% in pre‑market trading on the day of the announcement and a 3% year‑to‑date decline as of March 12, reflecting investor concerns about the company’s ability to keep pace with rivals and the potential impact on its AI‑driven revenue streams.

Meta’s AI portfolio also includes other projects such as Watermelon and the image/video generation system Mango, but the delay of Avocado underscores the challenges the company faces in delivering cutting‑edge models that can compete with industry leaders. The shift from an open‑source approach, exemplified by the Llama series, toward more proprietary development signals a strategic pivot aimed at protecting competitive advantages.

The content on EveryTicker is for informational purposes only and should not be construed as financial or investment advice. We are not financial advisors. Consult with a qualified professional before making any investment decisions. Any actions you take based on information from this site are solely at your own risk.