In December, reports revealed that Meta Platforms was developing a next-gen AI model, code-named Avocado, as a successor to its Llama series. Initially set for a first-quarter release, the launch is now postponed to May or June due to internal tests revealing Avocado’s performance lag behind competitors like Google’s Gemini 3.0 in areas like reasoning and coding. This shift marks a significant change in Meta’s approach—moving from open-source to a closed-source model aimed at increasing profit margins and maintaining exclusivity. However, the delay raises concerns about Meta’s hefty AI investments, with analysts questioning whether Avocado will still be competitive upon release. Despite these challenges, prioritizing quality over speed may ultimately benefit Meta, allowing for a robust model that can help reclaim a competitive edge in the fast-evolving AI landscape. This strategic pivot could safeguard investor confidence and enhance Metaverse’s long-term relevance.
Source link
Share
Read more