§ Documentation · v1

Build with Naxios.
Ship by lunchtime.

Guides, references, and runnable examples. Everything is OpenAI-SDK compatible — point base_url at Naxios and your existing clients keep working.

⌘K

Get started

Quickstart guides and core concepts

→

Models

Explore model capabilities and best practices

→

API reference

Complete API docs and endpoints

→

Examples

Real-world examples and code samples

→

Welcome to Naxios

Naxios provides the tools you need to build, test, and scale AI-powered applications with ease.

New to Naxios?

Start with the 5-minute quickstart to send your first API request and explore what's possible.

Top Tasks

1

Get your API key

Create an account and generate your API key.

→ 2

Make your first request

Send your first API request in minutes.

→ 3

Explore models

Find the right model for your use case.

→

Popular resources

Quickstart guide

Get up and running in 5 minutes.

Authentication

Secure your requests with API keys.

Model overview

Compare models and capabilities.

Rate limits

Understand usage limits and quotas.

Quickstart

Send your first request in three steps: grab an API key, install an SDK, and call a model.

1. Get an API key

Sign up and copy your key from the dashboard. Store it as an environment variable; never commit it to source control.

# .env NAXIOS_API_KEY=sk_live_your_key_here

2. Install the SDK

pip install naxios
npm install @naxios/sdk
# No install needed — cURL is built in

3. Make your first request

from naxious import NaxiosAI client = NaxiosAI(api_key=os.environ["NAXIOS_API_KEY"]) response = client.chat.create( model="naxios-pro", messages=[{"role": "user", "content": "Hello!"}] ) print(response.choices[0].message.content)
import { NaxiosAI } from "@naxios/sdk"; const client = new NaxiosAI({ apiKey: process.env.NAXIOS_API_KEY }); const res = await client.chat.create({ model: "naxios-pro", messages: [{ role: "user", content: "Hello!" }] }); console.log(res.choices[0].message.content);
curl https://naxios.tech/v1/chat/completions \ -H "Authorization: Bearer $NAXIOS_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "naxios-pro", "messages": [{"role": "user", "content": "Hello!"}] }'

Authentication

All requests must include a Bearer token in the Authorization header. Your API key authenticates you to the API and authorizes access to your account's resources.

Using your key

Authorization: Bearer sk_live_…

Best practices

  • Store keys in environment variables, secret managers, or your hosting provider's vault.
  • Never commit keys to git or expose them in client-side code.
  • Rotate keys periodically from your dashboard.
  • Use scoped keys for production vs. development where possible.

Error responses

Missing or invalid keys return 401 Unauthorized. Suspended accounts return 403 Forbidden.

Models

Naxios offers a catalog of text, vision, and audio models through one unified API. Pricing is the same across providers.

ModelBest forContext
naxios-proGeneral-purpose reasoning and coding200K
naxios-fastLow-latency conversational flows128K
naxios-thinkComplex multi-step reasoning200K
naxios-visionImage understanding and OCR128K
naxios-audioSpeech-to-text and synthesis—

Browse the full model catalog to see every supported model with provider logos and metadata.

API Reference

The API is REST over HTTPS, JSON in and JSON out, and is intentionally drop-in compatible with the OpenAI API — change the base URL and your key and most code will keep working.

Endpoints

POST /v1/chat/completionsChat and reasoning
POST /v1/completionsText completion
POST /v1/embeddingsVector embeddings
POST /v1/images/generateImage generation
POST /v1/audio/transcriptionsSpeech-to-text
GET  /v1/modelsList available models

Streaming

Pass stream: true to receive Server-Sent Events for token-by-token output.

const stream = await client.chat.create({ model: "naxios-pro", messages: [...], stream: true }); for await (const chunk of stream) { process.stdout.write(chunk.choices[0].delta.content || ""); }

Agents

Agents let a model call tools, reason over their output, and chain actions together. You describe the tools with JSON schemas and Naxios handles the orchestration loop.

{ "model": "naxios-pro", "messages": [...], "tools": [ { "type": "function", "function": { "name": "get_weather", "parameters": { /* JSON Schema */ } } } ] }

See the Guides for a step-by-step walkthrough of a research agent.

SDKs

Official SDKs wrap the REST API and add niceties like streaming iterators, retries with backoff, and typed responses.

  • Python — pip install naxios
  • JavaScript / TypeScript — npm install @naxios/sdk
  • Go — go get github.com/naxious-ai/naxios-go
  • HTTP — use any client with Bearer auth; see the API Reference

Guides

Recipe-style walkthroughs for common tasks. Each guide is self-contained and uses real, runnable code.

Rate Limits

Rate limits protect the platform and are enforced per API key. Your current tier and remaining quota are visible in the dashboard.

TierRPMTokens / day
Tier 0 — Starter5Free tier
Tier 1 — Adaptive201M
Tier 2 — Pro454M
Tier 3 — Premium9010M
Tier 4 — EnterpriseCustomCustom

When you exceed a limit the API returns 429 Too Many Requests with a Retry-After header. Back off, then retry.

Changelog

Product and API updates, newest first. Subscribe to the feed or follow us on Discord for real-time notifications.

  • Apr 23, 2026 — Dashboard v2 with usage charts and deterministic API keys.
  • Apr 10, 2026 — naxios-think reasoning model added.
  • Mar 28, 2026 — Streaming support for all chat endpoints.
  • Mar 15, 2026 — Free tier raised from 25K → 50K tokens/month.
  • Mar 1, 2026 — Vision endpoint naxios-vision launched.

Need help?

Join our community or contact support anytime.

Join Discord Contact support