Agents can now pay BlockRun per model request over x402, with wallet signing and spend limits handled by AWS infrastructure.
Incarna used Amazon Bedrock AgentCore payments to let its agents pay BlockRun for model inference one request at a time. BlockRun routes each request to more than 90 models from more than 15 providers over x402, and quotes and settles each call on its own.
Amazon Bedrock AgentCore gives teams a managed way to build and operate agents. AgentCore payments handles the payment protocol, connects to a wallet, signs the transaction, and enforces spending limits. Incarna provisions each agent wallet through the Coinbase CDP connector, and the customer keeps ownership while granting delegated authorization.
Card networks cannot handle the high-frequency, low-value pattern of paying per inference. AgentCore payments enforces limits at the infrastructure layer, so an agent cannot exceed them even when a prompt is manipulated. That moves payment controls away from model behavior and into the platform.
Payments settle in a stablecoin, and Incarna uses USDC on the Base network, which makes each transaction verifiable on-chain. The next signal to track is whether more inference routers and model providers adopt x402-compatible checkout on their endpoints.
What matters
- Incarna used AgentCore payments to let its agents pay BlockRun for model inference one request at a time.
- Builders add x402 payment support in days instead of months, with spending limits enforced outside the model.
- Watch whether AgentCore payments expands beyond x402 to reach more inference providers and networks.
Why it matters
Watch whether AgentCore payments expands beyond x402 to reach more inference providers and networks.
This GenAI News article was prepared in original wording using reporting and materials published by AWS Machine Learning Blog. Source reference: https://aws.amazon.com/blogs/machine-learning/pay-per-inference-for-ai-agents-how-blockrun-and-incarna-use-amazon-bedrock-agentcore-payments/.
Drafted by the GenAI News review pipeline.