Step 1: Get the request_id
Every response from the Pinaivu gateway includes a request_id field at the top level of the JSON response body. Capture it from any API call.
request_id — you will use it to retrieve the receipt in the next step.
Step 2: Look up the receipt
You have two options for retrieving a receipt: the REST API or the Explorer UI.- Option A — REST API
- Option B — Explorer UI
Send a A successful response looks like this:
GET request to the receipts endpoint, passing your request_id in the path and your API key in the Authorization header.Understanding the receipt fields
| Field | Description |
|---|---|
request_id | Unique identifier for the inference request, returned in every API response. |
node_id | The peer GPU node on the Pinaivu network that executed the inference. |
model | The model name and version that served the request (e.g., llama3.2:1b). |
timestamp | ISO 8601 UTC timestamp of when the inference was completed. |
payment_amount | The amount paid to the serving node for this request. |
coordinator_signature | Cryptographic signature produced by Pinaivu’s AWS Nitro Enclave, binding all fields above. |
Use cases for verification
- Auditing AI outputs — prove exactly which model and node produced a given response
- Cost tracking — reconcile per-request payment amounts against your usage budget
- Compliance reporting — attach routing receipts to AI-generated content as tamper-evident provenance records
- Debugging latency issues — identify which node served a slow request and correlate with node history in the Explorer
The signature is produced inside an AWS Nitro Enclave — an isolated, attested execution environment. This means the signature cannot be forged, even by Pinaivu itself.