Skip to main content
Pinaivu is a peer-to-peer marketplace for LLM inference that connects developers and applications with a global network of GPU node operators. Unlike traditional centralized providers, every request you send through Pinaivu is brokered by a cryptographically attested coordinator and backed by a signed, auditable routing receipt — so you always know what ran your inference and can prove it.

​
What is Pinaivu

When you send a request to Pinaivu, here is what happens at a high level: your request arrives at the OpenAI-compatible gateway at https://api.pinaivu.com/v1, the cryptographically attested coordinator auctions the job to competing GPU node operators, and the winning node runs your model and returns the result directly to your client. The coordinator then settles payment and issues a signed routing receipt tied to a unique request_id. From your application’s point of view, this looks identical to calling the OpenAI API. Under the hood, you get decentralization, cryptographic attestation, and a verifiable audit trail with every response.

​
Key features

OpenAI-Compatible API

The gateway at https://api.pinaivu.com/v1 speaks the OpenAI Chat Completions protocol. Drop in your Pinaivu key and switch models — no code changes required.

Verified Inference

Every completed request produces a signed routing receipt. You can cryptographically verify which node ran your inference and inspect the full proof.

Explorer

The public Explorer lets you look up any request_id to see node activity, payment settlement, and the signed routing receipt.

Chat Interface

chat.pinaivu.ai is a ready-made conversational front end for end users, with cross-session memory built in.

​
How it works

From your perspective as a developer, using Pinaivu involves four steps:
1

Send a request

Call POST https://api.pinaivu.com/v1/chat/completions with your API key and a model name, exactly as you would with the OpenAI API.
2

The coordinator auctions the job

Pinaivu’s attested coordinator selects a GPU node from the decentralized network based on availability and bid. This happens automatically — you do not need to configure anything.
3

Receive your response

The winning node runs the model and streams or returns the completion to your client. The response payload matches the OpenAI schema, including a Pinaivu-specific request_id field.
4

Verify on the explorer

Use the request_id from the response to look up the full routing receipt on explorer.pinaivu.com — including which node served the request and the signed proof of inference.

​
Where to go next

Quickstart

Create an account and make your first API call in under five minutes.

How It Works

Dive deeper into the coordinator, nodes, and the routing receipt model.