AI Price War Breaks Out: Meta Unveils Paid AI Model For First Time, Will Be "Among Most Affordable Options"

AI Price War Breaks Out: Meta Unveils Paid AI Model For First Time, Will Be "Among Most Affordable Options" Shortly after a leaked Meta memo revealed the company was planning on putting an AI chip into production in September as it looks to double computing capacity to 14Gigawatts, the company also unveiled a version of its most advanced artificial intelligence model, Muse Spark 1.1, that includes a new paid tier for developers , marking the first time Meta has charged businesses for access to its models and providing a new revenue stream. It’ll be " among the most affordable options " on the market, Zuckerberg said in a Bloomberg interview ahead of the release. “Since this is not an open source model, this is I think the first time that we’re doing a real serious API,” Zuckerberg said, referring to the application programming interface used to access Meta’s AI. “ And the pricing is going to be very aggressive and attractive ” he added indicating that Meta hopes to capture market share by undercutting its competitors. The new model’s biggest improvement is in its agentic capabilities, the Meta CEO told Bloomberg. He hopes to piggyback on the latest craze in AI development this year, which a month ago saw Goldman forecast that agentic AI use will lead to a massive 120 quadrillion monthly tokens being used by 2030. Agents are the big theme of AI this year, with the label applied to systems that can complete multistep tasks on behalf of a user. Zuckerberg described Muse Spark 1.1 as having “state-of-the-art or very close to it” agentic reasoning and tool use. The model is also greatly improved when it comes to coding and Meta employees are using it internally to build products and features for various apps, he added. Meta will also introduce a new Meta Model API system, which will be used to collect fees from developers. Its API pricing is roughly 25% of the cost advertised by other top models from OpenAI and Anthropic, according to Bloomberg. Developers will be able to use Meta’s model for free, but only up to a point; they’ll be required to pay for access after reaching a certain token threshold, Zuckerberg said. Which means that legacy frontier models will now have to worry about domestic cheap alternatives , especially after xAI also released an agentic and coding model yesterday which will have to grab market share, in addition to much cheaper Chinese models . “The pricing from some of the other labs is very extreme and has very high margins,” Zuckerberg said, underscoring that his strategy is to get Meta’s technology in front of as many people as possible. “ We think that there’s a real ability to be able to offer frontier or very high-level intelligence at a much more affordable cost .” Zuckerberg, 42, is spending aggressively to keep pace with rivals like OpenAI and Alphabet in a race to achieve what he calls superintelligence, or AI that can perform tasks better than humans. Meta has committed hundreds of billions of dollars to building the infrastructure necessary to develop superintelligence, including data centers and expensive AI chips. The company announced a new $10 billion data center investment in Canada as well as a new image-generation model just this week. Yet despite Meta’s massive investment spending, its models have not historically tested at the same level as those from Anthropic, OpenAI or Google. But Muse Spark 1.1 is more competitive, Zuckerberg said, and tested better than Google’s Gemini model in several categories related to agents, coding and multimodal capabilities. “That is a pretty interesting milestone because I think this may be the first time, at least that I can remember, that Meta’s models are better than all of the Google models,” he said, although it remains to be seen if users/Wall Street agree. Zuckerberg’s commitment to the AI race has led to a series of extreme shifts in strategy and resources over the past year. Following a disappointing model launch in the spring of 2025, Zuckerberg became personally involved in rebuilding Meta’s AI lab, which included hiring Scale AI’s Alexandr Wang to lead the new unit, and eventually, significant layoffs and several internal reorganizations. Once fully committed to building open source AI models that are available for free to outside developers, Meta has also pivoted toward prioritizing closed models that it can charge for — like Muse Spark 1.1 — which required essentially rebuilding the technologies from scratch. Zuckerberg said he’s pleased with how the lab is progressing, although it's not like he would openly admit the alternative. “We’re generally doing better than we expected,” Zuckerberg said. He acknowledged that Meta is still trailing some of the larger AI labs, including Anthropic and OpenAI, but said that the company has another new model coming, codenamed Watermelon, that he believes can help Meta “push this maximum frontier of intelligence.” He declined to share details on Watermelon’s release timeline, saying that the focus is on quality. Investing in a frontier model - or AI technology that moves the industry forward with new capabilities and features - is an expensive endeavor. But Zuckerberg believes it’s worth the investment from Meta given his mission: To build personal agents that everyone in the world can use. Which begs the question, posed earlier By Vital Knoweldge, who pointed out that just over the past 24 hours wee have seen "new frontier models from Meta (Muse Spark 1.1), SpaceX (Grok 4.5), and OpenAI (GPT 5.6), and asked "Seems like differences are fairly marginal. Does the world really need all these?" To Zuck the answer is, of course, yes: after all he has to justify the hundreds of billions he plans to sink into the commoditization of AI. Then again, Zuckerberg does not believe that the technology will ultimately become a commodity ( narrator: it will ) that is, that all of the various AI models will essentially do the same thing and be more or less indiscernible from one another . He pointed to Mythos, the latest model from Anthropic, which raised national security concerns in the US, as an example of how companies are already gatekeeping aspects of the technology instead of sharing it widely. The truth is that Zuck really has no choice: Meta is projecting record capital expenditures for 2026, in addition to spending billions on AI talent to build out its Meta Superintelligence Labs, and has pledged hundreds of billions more to infrastructure projects. Critics have questioned whether the pivot has paid off. One among them is Apollo chief economist Torsten Slok who in a note overnight wrotes that consensus expects free cash flow for the hyperscalers to more than double over the coming years. But what if the payoff takes longer than consensus assumes, Slok asks echoing a question that has plagued the AI industry since the summer of 2024? That question is particularly pressing given that token prices continue to decline and Chinese models are gaining ground, both in their share of the world's most-used models and in token usage, where they now lead their US counterparts among the top 20 models. PIggybacking on what we said a few weeks ago (see " Answering The "Trillion Dollar Question": Are China's AI Models A Better Value Than US Models "), Slok says that if Chinese models keep gaining and token prices keep falling, the hyperscaler cash flows expected may prove too optimistic. What are the consequences if the AI payoff comes slower than expected in the first chart? Cash flows and earnings disappoint : the projected free cash flow surge slips later while committed capex and heavy depreciation hit on schedule, squeezing margins and marking down the forecast in the first chart. A Mag 7 sell-off that takes the market with it: equity prices built on a fast payoff re-rate, and because the Magnificent 7 now account for so much of the indices, the pain can't stay contained, it spreads to chips, power, data centers and the S&P 500 as a whole. Balance sheets stretch and credit risk rises: with internal cash unable to cover spending, hyperscalers lean further on debt, raising leverage and inviting possible ratings downgrades if profits lag. All that is true, and is based on just growing Chinese competition. Now add various "new" domestic entrants such as xAI and Meta as aspirational frontier model leaders, which will inevitably spark an even more furious price war, and suddenly the return calculation for both stock and bond investors becomes much uglier. Tyler Durden Thu, 07/09/2026 - 11:05