JPMorgan Turns Bullish on Meta: Muse, Model API, and Subscriptions Support $820 Price Target

TL;DR
·JPMorgan upgraded Meta from Neutral to Overweight and raised its price target from $640 to $820, implying about 25% upside from the stock price at the time of the report.
·Muse Spark 1.3 has entered the frontier model tier, marking Meta's initial shift from a follower to a major competitor.
·The consumer AI agent Muse climbed to No. 3 on the U.S. App Store free chart the day after launch, with early user engagement 10 times that of internal test groups.
·Meta is expanding AI monetization beyond advertising into transaction commissions, paid subscriptions, enterprise agents, and model APIs, creating multiple new revenue streams.
·JPMorgan expects Meta's capital expenditures to reach $242.7 billion in 2027 and $283.5 billion in 2028, putting significant pressure on free cash flow.
·The report argues that current forecasts already incorporate most AI spending but have not yet included revenue from new products like Muse and model APIs, potentially underestimating Meta's earnings potential.

Why Does Meta Deserve a Re-rating?

Meta's stock has rebounded about 20% from its recent low, but it is still down 1% year-to-date, while the S&P 500 has risen about 12% over the same period. Amid ongoing market concerns about excessive AI spending and deteriorating free cash flow, JPMorgan upgraded Meta from Neutral to Overweight and raised its December 2027 price target from $640 to $820.

Based on the Sept. 9 stock price of $653.69, the new target implies about 25% upside. JPMorgan values Meta at approximately 23 times its 2028 GAAP earnings per share of $35.44, above the S&P 500's valuation of about 16 times.

The core judgment supporting this valuation premium is that Meta is transitioning from a social platform that uses AI to optimize recommendations and advertising to an AI platform that simultaneously offers frontier models, consumer agents, enterprise agents, developer APIs, and paid subscriptions. In the past, the market mainly saw capital expenditures; now Meta is beginning to show how these investments will translate into revenue.

Model capability is the foundation of this business system. When rebuilding Meta Superintelligence Labs in the summer of 2025, the company proposed launching a frontier-competitive model within one year. After releasing Muse Spark 1.1 in July 2026, Meta quickly iterated to Muse Spark 1.3.

According to third-party test results, Muse Spark 1.3 has entered the frontier model tier in multiple capabilities such as agents, coding, instruction following, and long-context, significantly narrowing the gap with leading models like Claude and GPT. JPMorgan therefore believes that Meta has essentially delivered on its previously stated model catch-up goal.

Artificial Analysis Model Intelligence Index. Muse Spark 1.3 has entered the frontier model tier and is competitive in capabilities such as agents, coding, and instruction following.

The next-generation model Watermelon is expected to further improve model capabilities. This model uses a higher level of pre-training, and the next-generation model after Watermelon has already begun scaled training on Meta's Prometheus gigawatt-scale computing cluster in Ohio.

Meta's real competitive advantage is not just the model itself, but the combination of models and distribution channels. The company's products cover about 4 billion users and connect hundreds of millions of businesses and advertisers. Once the underlying model reaches frontier levels, Meta can quickly embed it into Facebook, Instagram, WhatsApp, and Messenger, creating a scale advantage that other AI labs find difficult to replicate.

From Muse to Meta One, AI Begins to Find Monetization Entry Points

Muse is Meta's recently launched consumer AI agent, now available on iOS, Android, and the web in the United States. It can browse websites and operate user interfaces through its own virtual computer, completing tasks such as online shopping, restaurant reservations, filling out forms, and sending emails and messages on behalf of users.

Currently, Muse can work with platforms such as Instagram, WhatsApp, Spotify, DoorDash, Etsy, Reddit, Yelp, Outlook, and Gmail. The day after launch, it briefly climbed to No. 3 on the U.S. App Store free app chart, and early user engagement reached 10 times that of internal test groups, exceeding Meta's own expectations.

Muse's ranking on the U.S. App Store free app chart. Muse climbed to No. 3 on the U.S. App Store free app chart the day after launch, showing strong early user acquisition.

Muse currently offers free users a weekly quota of 100 million tokens, with two paid tiers at $20 and $100 per month. However, Meta does not plan to rely solely on subscription fees. Because Muse can directly shop, book, and complete transactions for users, the company is more likely to take commissions from transactions in the future.

This means Muse is not just targeting the chatbot subscription market, but the consumer transaction market, which could be worth tens of trillions of dollars. As automated interactions between agents increase, merchants may also gain more orders, customers, and business data through Muse.

Muse application scenarios and task execution flow. Muse uses a virtual computer to call different websites and applications, completing search, decision-making, and transactions on behalf of users.

Enterprise-side commercialization has progressed further. Meta Business Agent can help businesses answer questions, recommend products, book services, and screen sales leads. As of the second quarter of 2026, more than 1 million businesses per week were using this product on WhatsApp and Messenger.

The Business Agent Platform for large enterprises can connect to hundreds of external systems such as Shopify, Zendesk, and Shopee, allowing agents to execute actions directly on behalf of businesses. The platform officially launched token-based billing on August 1: $2 per 1 million tokens, with each message interaction costing about 4 to 5 cents.

In the future, Meta may also adopt a "pay-per-result" model similar to advertising auctions, allowing businesses to bid on actual outcomes such as sales, bookings, or lead conversions. Since Meta already has commercial relationships with hundreds of millions of advertisers and small and medium-sized businesses, Business Agent does not need to build an enterprise sales channel from scratch.

Meta Business Agent Platform pricing. Enterprise agents have started token-based billing, giving Meta a new entry point for enterprise service revenue.

The developer market is another monetization path. Meta Model API allows developers to call Muse Spark to build agents and multimodal workflows, while Muse Code is used for complex software engineering tasks.

The standard version of Muse Spark 1.3 is priced at $1.25 per million input tokens and $4.25 per million output tokens; the contributor version, which allows Meta to use data to improve products, has input and output prices of only $0.10 and $0.20, respectively. Muse Code also offers three subscription tiers at $5, $15, and $50 per month.

The low pricing indicates that Meta currently prioritizes usage scale and developer adoption. Early data already shows positive signals: the Muse Spark 1.3 contributor version ranks first in market share on OpenCode, having processed about 31 trillion tokens since launch, covering 212,000 unique users.

In addition to agents and APIs, Meta has also begun selling subscription services to individual users, businesses, and creators through Meta One. The individual plan is priced at $7.99 and $19.99 per month, offering higher Meta AI usage quotas and value-added features for Instagram, Facebook, and WhatsApp.

JPMorgan estimates that, based on an average annual revenue of $182 per individual user and a paid penetration rate of 2% to 3%, Meta One individual subscriptions could generate $14.2 billion to $21.3 billion in revenue and contribute $3.30 to $4.96 in GAAP earnings per share. If average annual revenue rises to $211, revenue could further reach $16.4 billion to $24.6 billion.

Meta One individual subscription revenue and EPS sensitivity analysis. Under a neutral penetration scenario of 2% to 3%, individual subscriptions could contribute approximately $14.2 billion to $24.6 billion in revenue.

Enterprise and creator versions are priced higher, ranging from $14.99 to $499.99 per month. Meta currently has more than 200 million business users and tens of millions of professional creators. Based on a neutral scenario of average annual revenue of $600 to $1,200 and penetration of 8% to 12%, this segment could contribute $12 billion to $36 billion in revenue in 2028, along with $2.80 to $8.40 in GAAP earnings per share.

Meta One enterprise and creator subscription revenue and EPS sensitivity analysis. Leveraging more than 200 million business users, enterprise and creator subscriptions could become a more flexible AI revenue source for Meta.

The above estimates do not yet include potential advertising revenue from Meta AI. In the future, Meta AI itself could become a new advertising display channel; user interaction data with the AI assistant could also improve ad targeting on Facebook, Instagram, and WhatsApp.

Massive Spending Is Not Over, but Revenue Is Not Yet Fully Priced In

Meta's AI commercialization path is becoming clearer, but that does not mean capital expenditure risks have disappeared. On the contrary, spending over the next two years may be even more aggressive.

JPMorgan expects Meta to maximize computing capacity expansion in 2026 and 2027. According to reports, the company plans to reach about 7GW of computing capacity in 2026 and double it to 14GW in 2027.

The report expects Meta's capital expenditures to rise from $69.7 billion in 2025 to $142.5 billion in 2026, then increase by 70% to $242.7 billion in 2027, and reach $283.5 billion in 2028, all significantly above market consensus.

Massive capital expenditures will directly pressure cash flow. JPMorgan expects Meta's free cash flow to decline from $47.1 billion in 2025 to negative $1.4 billion in 2026, and further to negative $58 billion and negative $49.6 billion in 2027 and 2028, respectively. The company may also shift from net cash to net debt in 2026, with net debt reaching about $165.2 billion in 2028.

Meta income statement and capital expenditure forecasts. Meta's capital expenditures will rise significantly over the next two years, expected to put continuous pressure on free cash flow.

However, this is also the key reason JPMorgan turned bullish on Meta: current financial forecasts already incorporate most AI infrastructure costs, but have not yet included incremental revenue from Muse, model APIs, and other new AI products. In other words, costs are already in the model, but potential revenue has not been fully calculated.

If Meta eventually ends up with excess computing capacity, the company could also rent out some of its computing power to external customers. Recently, some high-end computing contracts have been priced at $30 to $50 per watt, higher than the common $10 to $20 per watt for new cloud computing companies. However, JPMorgan expects Meta will still prioritize computing power for advertising optimization, frontier model training, and its own AI products, because these internal use cases may yield higher returns.

Meanwhile, the core advertising business remains the foundation for Meta to bear AI spending. AI can improve content recommendations, increase user time spent and ad targeting efficiency, and help advertisers generate creative materials at scale. JPMorgan expects Meta's revenue to grow from $201 billion in 2025 to $254.3 billion in 2026, and to $305.3 billion and $354.4 billion in 2027 and 2028, respectively.

The risks facing Meta remain clear: AI spending may continue to exceed expectations, and product monetization may not keep pace; Muse and other agents may fail to retain users over the long term; Google, TikTok, and OpenAI are also competing for user time and advertising budgets.

Therefore, whether the $820 price target can be realized depends not only on whether Meta can continue to invest in AI, but on whether Muse, Business Agent, model APIs, and Meta One can generate real revenue over the next two years. JPMorgan's judgment is that Meta has crossed the threshold of insufficient model capability, and AI spending is beginning to transform from a pure cost into a business system that can be subscribed to, called upon, charged for, and participate in transactions.