Naro

What the Naro API gives you

An API layer between your code and the model providers. It speaks the formats your SDKs already use, routes each request to the model you name, and meters every token.

Compatible with the SDKs you use

Works with the official openai packages for Python and Node.js, LangChain, LlamaIndex and the Vercel AI SDK. Change the base URL and key; request and response shapes stay the same.

POST /v1/chat/completions
OpenAI Chat Completions format, streaming, tools, JSON mode
POST /v1/embeddings
Text embeddings, OpenAI format
POST /v1/messages
Anthropic Messages format for Claude models
GET /v1/models
Live catalog with context window and pricing

stream.ts

import OpenAI from "openai";

const naro = new OpenAI({
  apiKey: process.env.NARO_API_KEY,
  baseURL: process.env.NARO_BASE_URL,
});

const stream = await naro.chat.completions.create({
  model: "google/gemini-2.5-pro",
  messages: [{ role: "user", content: "Draft a release note." }],
  stream: true,
});

Model catalog

Model IDs use the provider/model format. New provider releases are added to the catalog and listed by GET /v1/models.

OpenAI
gpt-4o, gpt-4o-mini, o3-mini, text-embedding-3-small, text-embedding-3-large
Anthropic
claude-sonnet-4, claude-3.5-haiku
Google
gemini-2.5-pro, gemini-2.5-flash
Meta
llama-3.3-70b-instruct, llama-3.1-8b-instruct
Mistral
mistral-large, mistral-small
DeepSeek
deepseek-v3, deepseek-r1

Routing and fallbacks

Add a models list and "route": "fallback". Naro tries each model in order on 429, 5xx or timeout, and bills only the request that succeeds.

request.json

{
  "model": "openai/gpt-4o",
  "models": ["anthropic/claude-sonnet-4", "google/gemini-2.5-pro"],
  "route": "fallback",
  "timeout_ms": 20000,
  "messages": [{ "role": "user", "content": "Classify this email." }]
}

Keys, limits and logs

Create keys per project or environment. Each key carries its own RPM and TPM limits, a monthly spend cap, an optional IP allowlist and a list of allowed models. Request logs show latency, token counts and cost for every call.

Illustrative product interface

See what each model costs.

View pricing