Official Source· AWS Artificial Intelligence· AI· Published Oct 8, 2026

AWS: AgentCore payments enables per-inference billing as Incarna goes live on Base

AWS describes how Incarna used Amazon Bedrock AgentCore payments and x402 to let AI agents pay BlockRun for each model-inference request. The integration took three days, and the flow is live with real settlement on Base mainnet.

AWS: AgentCore payments enables per-inference billing as Incarna goes live on Base

Full Report

In an article published October 8, 2026, AWS detailed how Incarna used Amazon Bedrock AgentCore payments to let its AI agents pay BlockRun for each inference request over x402. AWS describes AgentCore payments as a managed capability that handles payment protocols, wallet connections, transaction signing, and spending limits. It works with x402-compatible endpoints, including Amazon Bedrock inference endpoints.

BlockRun is a pay-as-you-go inference router offering more than 90 models from over 15 providers through x402. Incarna pays in USDC on Base, where transactions are verifiable on-chain. Each payment session has a spending ceiling and expiry; Incarna sets its sessions to match a day’s budget. AWS says the limits are enforced outside the model, so prompt manipulation cannot cause an agent to exceed them.

Incarna says its team completed the integration in three days—one day to build and two to test—with roughly 200 lines of application code. During beta, agents processed more than 1,000 payments, ranging from $0.001 to $0.05 per call, with each settling individually on-chain. AWS says AgentCore payments is generally available. The Incarna flow is live on Base mainnet, with BlockRun serving the sell side and AgentCore payments governing each transaction.

The example extends AgentCore’s documented use beyond areas such as enterprise cloud migration: earlier AWS coverage described a multi-agent migration framework built on AgentCore. For teams using agents with pay-per-use services, the Incarna case shows how individual payments, explicit spending ceilings, and on-chain verification can be incorporated into the workflow. AWS says the example reduced the work of adding x402 payment support from months to days.

Why This Matters

The case connects model inference to per-request payments and shows spending ceilings enforced outside the model. For teams relying on usage-based services, it offers a concrete example of the integration effort and transaction controls involved, rather than only describing agent payments as a concept.

Evidence

Start with the primary evidence, then review supporting sources and context.

1 items

Primary Evidence

1
  • OfficialAWS Artificial IntelligenceDocumentOriginal

    Pay-per-inference for AI agents: How BlockRun and Incarna use Amazon Bedrock AgentCore payments

    Peter Jiang

    “Amazon Bedrock AgentCore payments gives AI agents a managed way to pay for services on demand, with spending limits enforced by the infrastructure. See how Incarna's agents pay BlockRun for model inference one request at a time over x402, cutting the work of adding x402 payment support from months to days.”

Recommended signals

Recommended by brand, topic, and recent related coverage

No more recommended signals right now

Official statement

Open original source