Meta (META) shares experienced a 4.90% decline on Friday, despite encouraging signs of advancement in the company’s artificial intelligence initiatives. During an internal town hall meeting, Alexandr Wang, who heads the company’s superintelligence division, informed staff members that Meta’s forthcoming AI model, internally referred to as Watermelon, has now achieved performance levels comparable to OpenAI’s leading GPT-5.5 system — a development that could represent a significant achievement in the social media giant’s AI ambitions.
Wang referenced various AI model performance metrics to support his assertion, although the specific benchmarks used in the comparison remain undisclosed.
The Watermelon model represents Meta’s latest advancement following Avocado — the code name assigned to Muse Spark, which debuted in April. While Muse Spark demonstrated competitive benchmark performance upon release, it remained behind leading models from OpenAI and Anthropic.
According to Wang, Watermelon employs “an order of magnitude” more computational resources than Avocado. While increased computing power generally yields more sophisticated capabilities, it also substantially increases operational expenses.
Wang also shared some progress updates publicly. Through a post on X, he announced that an upgraded version of the existing Muse Spark model would arrive shortly, featuring substantial enhancements in programming abilities and autonomous agent functionality.
Responding to a question from a user about when Meta might deliver a coding-focused model comparable to Anthropic’s Claude Opus, Wang indicated it would happen “pretty soon” and suggested users would appreciate what the company is developing.
Meta has invested considerable time and resources attempting to narrow the performance gap with OpenAI, Google, and Anthropic. Massive expenditures on processing chips, computing facilities, and premier AI researchers have yet to establish the company as an industry leader — at least not in publicly available products.
Zuckerberg brought Wang aboard last year to spearhead what was restructured as Meta Superintelligence Labs. Wang directs an elite research group known internally as TBD, alongside managing the company’s comprehensive AI and hardware initiatives.
Meta has extended compensation packages worth hundreds of millions of dollars to attract leading AI researchers. The financial commitments extend well beyond talent acquisition.
The company informed shareholders it now anticipates capital expenditures between $125 billion and $145 billion in 2026 for processors, computing facilities, and related infrastructure. This represents an increase from the earlier projection of $115–$135 billion, attributed to escalating component prices and expanded data center construction.
For perspective, OpenAI launched GPT-5.5 in April. The organization subsequently introduced GPT-5.6 late last month — characterized as its most capable model to date — though widespread access has been restricted, allegedly at the behest of US government officials.
Wang’s assertions have not undergone independent verification, and Meta representatives declined to provide comment. OpenAI did not respond to inquiries regarding the matter.
The post Meta (META) Stock Drops 4.9% Despite Internal AI Breakthrough Matching OpenAI appeared first on Blockonomi.
