# Nirvani AI

> Nirvani AI is one API for leading AI models, built by Nirvani Company (Scottsdale, Arizona). It follows the standard chat-completions format and adds production controls: streaming, per-key rate limits, monthly token quotas and live usage metering. Base URL: https://api.nirvani.ai/v1. Free test keys are delivered by email at https://ask.nirvani.ai/get-key.

Key facts:
- Authentication: `Authorization: Bearer nvk_test_...` or `nvk_live_...`. Keys are shown once by email and stored only as SHA-256 hashes.
- Models: `nirvani-flash` (fast default), `nirvani-flash-lite` (lowest cost), `nirvani-pro` (deepest reasoning and writing), `nirvani-open` (open weights). Live list: `GET https://api.nirvani.ai/v1/models` (no key needed).
- Test keys: 20 requests per minute and 100,000 tokens per month. Live keys: 120 requests per minute and 5,000,000 tokens per month, set up per business.
- Existing chat-completions SDKs work after setting the base URL to https://api.nirvani.ai/v1.
- Nirvani Company also provides AI revenue infrastructure for small and mid-sized businesses: titled AI agents such as the AI Receptionist, AI Live Voice Agent, AI Live Chat Agent, AI Sales Rep and AI Follow-Up Specialist. See https://nirvani.ai.

## Docs

- [API documentation](https://ask.nirvani.ai/docs/): quickstart, authentication, models, chat completions, streaming, usage, limits, errors
- [Full documentation as plain text](https://ask.nirvani.ai/llms-full.txt): the complete reference in one file
- [Models endpoint](https://api.nirvani.ai/v1/models): live model registry as JSON

## Product

- [Chat](https://ask.nirvani.ai/chat/): talk to Nirvani AI in the browser, no sign-up
- [Playground](https://ask.nirvani.ai/playground/): compare models side by side with live latency
- [Console](https://ask.nirvani.ai/console/): usage, latency, errors and quota per key
- [Get a free key](https://ask.nirvani.ai/get-key/): free test key by email

## Optional

- [Nirvani](https://nirvani.ai): AI agents for small and mid-sized businesses