On Wednesday, Meta Platforms (META) officially launched its new AI model, “Muse Spark,” marking the first core achievement since its heavy investment in AI. This move signals the company’s attempt to regain a voice in a market dominated by OpenAI, Anthropic, and Google.
The model was developed by Meta’s “Superintelligence Lab,” led by former Scale AI CEO Alexandr Wang. Unlike “flagship models” that pursue极致 performance, Muse Spark focuses on being “lightweight and efficient,” offering strong reasoning capabilities in complex tasks such as science, mathematics, and healthcare, while significantly reducing computing costs. Meta stated that by restructuring its AI infrastructure, the new model achieves performance close to that of its previous Llama 4 mid-sized model with lower computational resources.
Boosted by the positive news, Meta’s stock surged nearly 9% during the day, marking its largest single-day gain since January this year, before closing up 6.5% at $612.42. Strategically, Meta is accelerating its AI commercialization. On one hand, the company plans to open Muse Spark to developers via API to explore new revenue streams. On the other hand, the model will be gradually integrated into core product ecosystems such as Facebook, Instagram, WhatsApp, and Ray-Ban smart glasses.
Notably, Meta is significantly increasing its AI investments. The company expects AI-related capital expenditures to reach $115 billion to $135 billion in 2026, nearly double the previous year. Meanwhile, the global generative AI market is projected to grow from approximately $22 billion in 2025 to $325 billion by 2033, with a compound annual growth rate exceeding 40%.
Previously, Meta invested substantial resources in its Llama large language model, which is widely considered to lag behind industry leaders. CEO Mark Zuckerberg has repeatedly staked the company’s future on artificial intelligence, pouring tens of billions of dollars into the Reality Labs division. The release of Muse Spark is seen as a major counterstrike by Meta in the AI field. While it is still too early to assess the specific impact of the model, investor enthusiasm has already been fully demonstrated.
