Meta's 'Avocado' AI model delayed over performance issues
What's the story
Meta has pushed back the release of its artificial intelligence (AI) model, codenamed "Avocado," to at least May, according to The New York Times. The delay comes despite the company's heavy investments in expanding its AI capabilities and plans to develop its own chips. The new model's performance is said to be between Google's Gemini 2.5 and Gemini 3, which has contributed to the postponement of its launch.
Model development
Model's performance not on par with competitors
Meta has been working on the Avocado model for months. However, it hasn't performed as well as the latest offerings from competitors. A Meta spokesperson told Reuters, Our next model will be good, but more importantly, show the rapid trajectory we're on. They added that the company plans to steadily push the frontier over the course of the year as we continue to release new models.
Investment strategy
Spending plans for AI superintelligence
In January, Meta unveiled its capital spending plans for the year, estimating between $115 billion and $135 billion. The company's goal is to achieve "superintelligence," a stage where AI surpasses human intelligence. Despite the delay in Avocado's release, Meta remains committed to its ambitious vision and continues to invest heavily in this area.
AI licensing
Licensing Gemini AI for temporary boost under consideration
The leaders of Meta's AI division have discussed the possibility of temporarily licensing Google's Gemini to power their AI products. However, no decisions have been made yet. This potential collaboration highlights the competitive nature of the tech industry and how companies are willing to explore partnerships for technological advancement.