The peer-to-peer market for AI inference.
Anyone can sell inference for any model, at their own price. Buyers and agents pay per request, and each request goes to the cheapest healthy seller on Inferit.
| Buy | Sell | |
|---|---|---|
| Start | An API key from the Buy workspace, or x402 per request from a wallet, with no account and no API key | One command on your own computer, with a token from Add a machine in the Sell workspace |
| What you run | Any OpenAI-compatible client: change the base URL and the key | Ollama, LM Studio, vLLM or llama.cpp. The app dials out, so you need no open port and no domain |
| Money | Per-token prices, charged per request in µ (micro-dollars). A key spends your balance under a ceiling you sign, with an expiry the contract enforces | Your own price per 1M tokens, paid per request to the payout wallet you register (test ITC on the testnet) |
| Guide | Quickstart · Agents and x402 | Sell from your own PC |
Create a key in the Buy workspace and keep it in your environment as INFERIT_API_KEY.
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api-production-c74b9.up.railway.app/v1",
api_key=os.environ["INFERIT_API_KEY"],
)
response = client.chat.completions.create(
model="qwen/qwen-2.5-14b-instruct",
messages=[{"role": "user", "content": "Explain a spending cap in one sentence."}],
)
print(response.choices[0].message.content)Name a model with a live offer (inferit.app/models), or send model="auto" and Inferit picks the model. Auto chooses among ranked models with a healthy seller that fit the request, weighing a published quality ranking (Artificial Analysis Intelligence Index, read 2026-10-06) against this request's price. It never reads the prompt, and the x-inferit-model header names the model that answered. When no ranked model has a healthy seller, auto answers 503 no_eligible_model and nothing is charged; a named model keeps working. The method is on inferit.cc/docs/routing.
An agent can skip the key: a call with no key gets 402 Payment Required with a quote; the agent signs the payment with an x402 v2 client and resends the same request. See inferit.cc/agents.md.
npx -y inferit-seller connect --token icn_… && npx -y inferit-seller install-serviceThe app finds your local model server, matches its models to the catalog and runs as a background service that restarts after a reboot. Set your prices in the Sell workspace.
| Repository | What it is |
|---|---|
| inferit-seller | Sell inference from your own computer with one command. Apache-2.0, on npm as inferit-seller |
| whitechain-x402-facilitator | The first public x402 v2 facilitator for Whitechain, live on Whitechain Sepolia (testnet). Apache-2.0 |
| inferit-examples | Runnable examples for LangChain, the Vercel AI SDK, LiteLLM, CrewAI and OpenCode, paying with an API key or x402 |
| inferit-agent-example | A minimal TypeScript agent that pays Inferit per request with x402: no account and no API key |
| model-catalog | The models the Inferit API knows, as open data: ids, aliases, context windows and licence classes (CC BY 4.0) |
| awesome-inferit | Docs, guides, repositories and endpoints for Inferit, in one index |
| awesome-x402-whitechain | x402 resources for Whitechain: the spec, SDKs, EIP-3009, the testnet, facilitators and examples |
| branding | The Inferit logo files, colours, type and usage rules |
Inferit runs the market API and the router; their code is private.
Status: Testnet demo on Whitechain Sepolia. Prices settle in Inferit Test Credit (ITC), a test token with no monetary value. The contracts are not audited. Inferit has no token.
Website · App · Docs · API reference · X