Pay-per-inference for AI agents: How BlockRun and Incarna use Amazon Bedrock AgentCore payments
In this post, we look at how Incarna used AgentCore payments to let its agents pay BlockRun for model inference one request at a time.
Proof1 independent outlet
Key points
- When an AI agent runs, it often needs to buy something to finish a task: a model inference, an API response, access web content, or a call to another agent.
- BlockRun is a pay-as-you-go inference router serving more than 90 models from more than 15 providers over x402, each call quoted and settled independently.
- AgentCore payments works with x402-compatible endpoints, including Amazon Bedrock inference endpoints.
- The challenge: Paying for inference by the request
Sources (1)
- [1]Pay-per-inference for AI agents: How BlockRun and Incarna use Amazon Bedrock AgentCore paymentsAWS Machine Learning Blog · Oct 8, 06:33 PM
In this post, we look at how Incarna used AgentCore payments to let its agents pay BlockRun for model inference one request at a time.
When an AI agent runs, it often needs to buy something to finish a task: a model inference, an API response, access web content, or a call to another agent.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 8, 2026Can AI Agents Learn Their Way to the Top? Evaluating Heuristic Learning in a Long-Running Game Agent Competition
- Oct 7, 2026[AINews] Claude Haiku 5.5 — better than GPT-6 Luna at the same pricing
- Oct 7, 2026Show HN: Agent.reviews – Where AI agents read and write reviews on tools
- Oct 7, 2026Agent Lightning v1.0: A 3,500-Line Lightweight Agentic RL Framework for Training Agents with Real Harnesses
- Oct 7, 2026Building AI builders: Playbook for closing the AI knowledge-capability gap
- Oct 7, 2026AgentTime: Can Agents Estimate and Control Their Own Runtime?
