Tempo and Stripe's MPP Protocol: Enabling AI Agents to Pay-Per-Call APIs
Tempo's official blog details API monetization models and proposes the MPP protocol, which lets AI agents pay directly at request time—bypassing account opening and subscriptions—and provides a machine-readable payment path for per-call billing.

In an official blog post published on October 24, 2026, Tempo systematically reviewed four mainstream API monetization models—subscriptions, usage-based billing, prepaid credits, and pay-per-request—and highlighted the Machine Payments Protocol (MPP) it co-drafted with Stripe, which is designed to let AI agents complete payment directly when calling APIs.
According to the blog, MPP works over HTTP: a client calls a paid endpoint, the API returns HTTP 402 Payment Required along with payment terms, the client authorizes payment from a compatible wallet, and the API verifies it and returns the response plus a receipt. For high-frequency calls, MPP supports a session mode that allows charges to accumulate and settle periodically rather than paying separately for each request.
This design targets the distinctive behavior of AI agents as API consumers. The blog notes that agents dynamically discover and call APIs during tasks and typically do not register accounts, bind bank cards, or copy API keys for each new provider; their call volumes are small and bursty, and they may call only a few times and never return; at the same time, agents need to read prices at runtime and make decisions within budget. Traditional subscription or prepaid credit models struggle to fit these characteristics, whereas pay-per-request—if it can be low-cost and machine-readable—becomes the natural choice.
From the perspective of agent payment compliance and security that OceanAlt focuses on, MPP embeds payment decisions into the request path, meaning the intervention point for pre-settlement firewalls moves earlier. When an agent autonomously authorizes payment after an HTTP 402 response, judgments about who is paying, to whom, and whether this payment should be made need to be completed at millisecond latency, and binding KYA (Know Your Agent) identity verification with authorization intent (mandate) becomes an issue the protocol layer must consider. The blog does not mention specific mechanisms in MPP for identity verification or compliance screening, but per-call billing naturally produces high-frequency, small-value, cross-provider payment flows, which imposes new requirements on the granularity of anti-money laundering and sanctions list screening.
For API providers, MPP offers a path to serve "one-time callers" without bearing the overhead of account systems and subscription management; for agent developers, machine-readable pricing and payment terms lower integration costs. Whether the protocol can be widely adopted, however, depends on wallet compatibility, credit risk within the settlement cycle, and regulators' attitudes toward autonomous agent payments. The background of the Tempo-Stripe collaboration is noteworthy: Stripe previously acquired stablecoin infrastructure provider Bridge, and co-drafting MPP this time continues its positioning in machine-to-machine payments.
Original source: Tempo official blog · https://tempo.xyz/learn/api-monetization/
Provenance & status
- Byline
- OceanAlt Editorial
- First published
- 2026-10-27
- Last updated
- 2026-10-27
- Content type
- Original compilation
- Source material
- View original ↗
Related reading
Paste a payee address before you pay and see whether it's on a sanctions list, through a mixer, or tagged for fraud.
This judgement can sit inside your own product
One line of code; it touches neither your CSS nor your JS. The same pre-settlement judgement can appear in your articles, on your wallet's confirmation screen, or as an endpoint your agent calls before it pays.



