Meta Platforms on Thursday released developer access to its Muse Spark AI model and announced plans to manufacture its own artificial intelligence chip from September, marking the social media giant's most aggressive push yet to reduce dependence on Nvidia and cut its ballooning AI infrastructure costs.
The company released Muse Spark 1.1, which it describes as its most capable model for real-world coding and agentic tasks, alongside a pricing structure that puts it in direct competition with Anthropic and OpenAI. Developers in the United States can now access the model through Meta Model API, receiving $20 in free credits before switching to pay-as-you-go pricing of $1.25 per million input tokens and $4.25 per million output tokens.
The Fiscal Reality
Meta expects to spend as much as $145 billion on AI infrastructure this year, a significant portion of Big Tech's more than $700 billion projected outlay on the technology. That staggering figure explains why the company's pushing ahead with its own chip production. The custom chip, code-named "Iris," is part of a four-generation project for Meta Training and Inference Accelerators designed in-house to reduce costs and gain independence from suppliers like Nvidia and Advanced Micro Devices.
An internal memo reviewed by Reuters said testing the chip took only six weeks and found no major issues. Meta's working with Broadcom to help design it and Taiwan Semiconductor Manufacturing Co to manufacture it. The approach is likely to help the firm lower its massive computing costs, though adopting the latest GPUs at a firm as large as Meta "has been a heavy lift, and it has cost us time," the memo showed.
The upgraded Muse Spark 1.1 model can write and debug code, use software and external tools, understand text, images and video, and carry out complex multi-step tasks with less human intervention. Meta CEO Mark Zuckerberg said in a post on X, "Our focus is on delivering strong agentic and multimodal models at very low cost."
Infrastructure Expansion
This year, Meta plans to deploy seven gigawatts of computing infrastructure. The company added one gigawatt in the first half of the year and forecasts adding another 5.5 gigawatts by the end of the year, according to the memo. One gigawatt of energy is enough to power about 800,000 homes. Meta plans to double capacity again next year to reach a total of 14 gigawatts in 2027.
To expand computing infrastructure, Meta's secured long-term, multi-year supply agreements with Samsung Electronics for memory chips, Sandisk for flash storage and Sumitomo Electric for fiber-optic equipment. Components such as memory and AI chips have experienced a surge in demand as tech companies race to expand data centers to keep pace with AI's thirst for computing power.
Memory and other chip prices have risen rapidly and substantially enough that "chipflation" has become a macroeconomic concern, Morgan Stanley analysts said. That's precisely why Meta's betting on vertical integration through its own chip production.
Market Positioning
Meta's pricing sits above OpenAI's entry-level GPT-5 mini and Anthropic's low-cost Claude Haiku 4.5, but below Anthropic's higher-end Claude Sonnet 4.6 model. The new model is now available in Thinking mode in the Meta AI app and on the website. It's also expected to replace existing Llama models powering chatbots on WhatsApp, Instagram, Facebook and Meta's collection of smart glasses.
In April, Meta debuted Muse Spark, the first text and reasoning AI model from the superintelligence team it assembled last year to close the gap with rivals in the heated competition for AI supremacy. Meta was testing the Application Programming Interface with partners in a private preview during its launch. The API is a key element for AI systems, acting like a digital bridge for developers that allows them to use the model's capabilities in their own software systems.
The release follows a company announcement on Tuesday expanding generative AI tools across its apps by rolling out Muse Image, its first image-generation model from Meta Superintelligence Labs. Meta plans to launch a chip about every six months through 2027, whereas typically firms release AI chips at intervals of a year or more. The chip is meant to augment the large quantities of graphics processing units used for AI applications that Meta purchases from Nvidia and AMD.
Why This Matters:
Meta's dual announcement reveals the economic reality driving Big Tech's AI strategy: the cost of relying on external chip suppliers has become unsustainable. With $145 billion earmarked for AI infrastructure this year alone, Meta's vertical integration through its own chip production isn't just a technical decision — it's a financial imperative. The company's aggressive timeline, launching a new chip every six months through 2027, shows how seriously it's taking the cost problem. For European policymakers watching American tech giants dominate AI development, this underscores the competitive disadvantage of Europe's fragmented approach. The scale of investment required — 14 gigawatts of computing capacity by 2027, enough to power roughly 11 million homes — also raises questions about energy policy and whether Europe's regulatory environment can support similar infrastructure buildouts. The "chipflation" concern flagged by Morgan Stanley analysts isn't just a tech sector issue; it's a macroeconomic risk that could ripple through supply chains and consumer prices.