OpenAI-compatible gateway

One API for
every model.

Neviri AI routes one OpenAI-compatible endpoint to frontier and open models — resold and self-hosted. Sign in with your Neviri account, add credits, and ship.

drop-in · openai sdk
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.neviri.ai/v1",
  apiKey: process.env.NEVIRI_AI_KEY,   // nvai_...
});

const res = await client.chat.completions.create({
  model: "neviri/flagship-chat",
  messages: [{ role: "user", content: "Hello, Neviri." }],
  stream: true,
});

Why Neviri AI

One endpoint

An OpenAI-compatible /v1 API. Swap models by changing one string — no rewrite.

Resold + self-hosted

Frontier models from upstream providers and open models on Neviri GPUs, routed per model.

Usage-based

Pay for the tokens you use. Per-key spend caps and live metering keep spend in view.

Your Neviri login

Sign in with the account you already have. One session across Neviri Cloud and Neviri AI.

Explore models

One key reaches all of them.sample catalogue

Browse all

Flagship Chat

Resold
Text·128K ctx
price / 1M tokens

Reasoning Pro

Resold
Text·200K ctx
price / 1M tokens

Neviri Open 8B

Self-hosted
Text·32K ctx
price / 1M tokens

Vision Multimodal

Resold
Text · Image·128K ctx
price / 1M tokens

Embed Large

Self-hosted
Embeddings·8K ctx
price / 1M tokens

Fast Mini

Resold
Text·64K ctx
price / 1M tokens

From nothing to your first token in three steps.

01

Sign in with Neviri

Use your existing Neviri account — no new signup. Your session carries over from the console.

02

Add credits

Top up your Neviri AI wallet — a signup credit gets your first requests moving.

03

Call the API

Create a key, point any OpenAI SDK at Neviri AI, and start streaming completions.

Ship your first request today.

Sign in with Neviri, grab a key, and call any model from one API.