Let your AI agent check GPU availability, pricing and pods on RunPod during a call.

RunPod is a GPU cloud for AI and machine learning workloads. Once you connect your RunPod account to AgentCalls, your voice agent can look up available GPU types and their pricing, check the status and configuration of a pod, and list CPU options. Callers get accurate answers about compute without waiting for someone to log in. Your team can also have the agent create secrets or clusters for follow-up work. You connect RunPod with your own credentials from its settings.

What your agent can do with RunPod

  • Look up available GPU types, specifications and current pricing during a call
  • Check the status, GPU count, memory and cost of a specific pod
  • List CPU types and options for provisioning
  • Create secure secrets for credentials your pods need at runtime
  • Create a multi-node GPU cluster from a call request, with your confirmation

RunPod exposes 13 actions. You choose which ones the agent may use and whether each runs during or after the call.

Example calls

Quoting GPU options for a prospect

  1. Caller

    Hi, I need to train a small vision model next week. What GPUs can I rent right now?

  2. Agent

    Sure. Are you looking for something under a certain budget, or do you need a specific amount of memory?

  3. Caller

    Around 24 GB of memory, and I want to keep it under a dollar an hour.

  4. Agent

    Two options fit that today. One has 24 GB at a lower hourly rate, and the other has more memory at a slightly higher rate. I can send you both with the current prices.

Checking on a running pod

  1. Caller

    This is Priya from Northwind Labs. Can you check pod 7f3a9c on our account?

  2. Agent

    Thanks, Priya. Pod 7f3a9c is running with one GPU and 32 GB of memory. It has been up since Tuesday at 4:10 pm.

  3. Caller

    Great, and what is it costing us per hour?

  4. Agent

    It is billed at $0.79 per hour. Would you like me to leave a note for the team that you checked on it?

Outbound follow-up on a finished job

  1. Agent

    Hello, this is the Brightpath team calling for Marcus. Your training run finished overnight.

  2. Caller

    Oh good. Did it use the full budget?

  3. Agent

    It used about 62 percent of the planned hours, so the total came to $41.30. Do you want the pod kept running for another run?

  4. Caller

    No, shut it down after the results are saved.

How to connect RunPod

  1. 1

    Create your agent

    Sign up to AgentCalls, pick an English or Arabic voice and write what the agent should know and do in plain language.

  2. 2

    Connect RunPod

    From Integrations in the dashboard, choose RunPod and enter the connection details from your RunPod account settings.

  3. 3

    Choose the actions

    In the agent editor, enable the RunPod actions the agent may use and set each to run during the call or afterwards. Then test it from your browser.

Questions, answered.

Does the agent work with RunPod during the call or after it?

During the call. Your agent can look up GPU types, pod details and pricing while the caller is still on the line. It only creates or changes resources when the request is clear and confirmed.

What do I need to connect RunPod?

A RunPod account and its credentials. Enter them in the RunPod settings from your AgentCalls dashboard. You can disconnect at any time.

Which plan includes this integration?

Integrations are included from the Growth plan. Connect RunPod from your dashboard once your plan is active.

Page drafted from the tool's public description and reviewed automatically for accuracy of claims

Put your agent to work with RunPod.

Connect in minutes, test from your browser, go live on a number in the US, UK or Gulf. Free to start, no card needed.