Make your first request

Generate your setup

Pick your tool and model, then copy the exact settings. Only models that support the API your tool uses are listed.

export ANTHROPIC_BASE_URL='https://api.nightrouter.ai'
export ANTHROPIC_AUTH_TOKEN='sk-nr-your-key'
export ANTHROPIC_MODEL='claude-fable-5'

Authentication

Send your key as a Bearer token in the Authorization header. Anthropic-style clients may send it in the x-api-key header instead.

Send a request

A minimal call to the chat completions endpoint. Put your own key in place of the placeholder.

curl https://api.nightrouter.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-nr-your-key" -H "Content-Type: application/json" \
  -d '{"model":"claude-fable-5","messages":[{"role":"user","content":"Hello"}]}'

Models & pricing

Use a model id from the model library. Prices are listed per 1M tokens or per request, and your usage is recorded in the logs of your dashboard.

Claude Code

Claude Code reads its endpoint and key from environment variables. Add these lines to your shell profile, then run claude.

export ANTHROPIC_BASE_URL='https://api.nightrouter.ai'
export ANTHROPIC_AUTH_TOKEN='sk-nr-your-key'
export ANTHROPIC_MODEL='claude-fable-5'

Codex CLI

Codex CLI talks to the Responses API. Add this provider to its config file and export your key as NIGHTROUTER_API_KEY.

model = "your-model-id"
model_provider = "nightrouter"

[model_providers.nightrouter]
name = "NightRouter"
base_url = "https://api.nightrouter.ai/v1"
env_key = "NIGHTROUTER_API_KEY"
wire_api = "responses"

# export NIGHTROUTER_API_KEY="sk-nr-your-key"

Coding agents

In tools such as Cline, Roo Code or Cursor, choose the OpenAI-compatible provider option, set the base URL to https://api.nightrouter.ai/v1 and paste your key.

API Provider  OpenAI Compatible
Base URL      https://api.nightrouter.ai/v1
API Key       sk-nr-your-key
Model ID      claude-fable-5

OpenAI API Key            sk-nr-your-key
Override OpenAI Base URL  https://api.nightrouter.ai/v1
Add custom model          claude-fable-5

The same base URL and key may also work in other tools that let you set a custom OpenAI-compatible base URL, for example Kilo Code, Continue and Zed.

SDKs

Point the official SDKs at NightRouter by changing the base URL and the key.

from openai import OpenAI

client = OpenAI(base_url="https://api.nightrouter.ai/v1", api_key="sk-nr-your-key")
r = client.chat.completions.create(model="claude-fable-5", messages=[{"role": "user", "content": "Hello"}])
print(r.choices[0].message.content)

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({ baseURL: "https://api.nightrouter.ai", apiKey: "sk-nr-your-key" });
const msg = await client.messages.create({ model: "claude-fable-5", max_tokens: 1024,
  messages: [{ role: "user", content: "Hello" }] });

Chat clients

Chat apps that let you set a custom OpenAI-compatible base URL, for example Cherry Studio, LobeChat, Open WebUI, LibreChat or TypingMind, can be pointed at NightRouter: add a custom provider with the base URL https://api.nightrouter.ai/v1 and your key.

API reference

These endpoints live under https://api.nightrouter.ai and use the same key.

curl https://api.nightrouter.ai/v1/responses \
  -H "Authorization: Bearer sk-nr-your-key" \
  -H "Content-Type: application/json" \
  -d '{"model":"your-model-id","input":"Hello"}'

curl https://api.nightrouter.ai/v1/messages \
  -H "x-api-key: sk-nr-your-key" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{"model":"claude-fable-5","max_tokens":1024,"messages":[{"role":"user","content":"Hello"}]}'

Errors & limits

Errors use standard HTTP status codes, and the response body usually includes an error message.

These are general descriptions of common errors, not a service-level commitment.