API Reference

ClosedRouter is a fully OpenAI-compatible proxy. Point your existing SDK or HTTP client at our endpoint and route between every major AI provider through a single API key.

Quickstart

Get an API key

Make your first request

curl https://api.closedrouter.net/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-4o-mini",
    "messages": [
      {"role": "user", "content": "Hello! What model am I talking to?"}
    ]
  }'

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "model": "openai/gpt-4o-mini",
  "choices": [{
    "message": {
      "role": "assistant",
      "content": "You're talking to GPT-4o-mini, routed through ClosedRouter!"
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 14,
    "total_tokens": 26
  }
}

Swap providers instantly

Change the model field to route to any provider. Same request shape, different model:

# Google Gemini
curl https://api.closedrouter.net/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "google/gemini-2.5-flash", "messages": [{"role": "user", "content": "Hi"}]}'

# xAI Grok
curl https://api.closedrouter.net/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "x-ai/grok-3", "messages": [{"role": "user", "content": "Hi"}]}'

# Zhipu GLM
curl https://api.closedrouter.net/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "z-ai/glm-5.1", "messages": [{"role": "user", "content": "Hi"}]}'

Authentication

Include your API key in the Authorization header as a Bearer token:

Authorization: Bearer sk-xxxxxxxxxxxxxxxx

Keep your key secret. Never expose it in client-side code. Use environment variables or a secrets manager.

Models

Models are identified by provider/model-name. Browse all available models at closedrouter.net/models.

Model ID	Provider	Context
`openai/gpt-4o-mini`	OpenAI	128K
`google/gemini-2.5-flash`	Google	1M
`x-ai/grok-3`	xAI	131K
`z-ai/glm-5.1`	Zhipu	128K
`deepseek/deepseek-chat-v3.1`	DeepSeek	128K
`moonshotai/kimi-k2.5`	Moonshot	128K

List models

curl https://api.closedrouter.net/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"

Chat completions

POST /v1/chat/completions

Drop-in replacement for the OpenAI chat completions endpoint. Supports all standard parameters.

Parameter	Type	Description
`model`	string	Required. Model ID in provider/model format.
`messages`	array	Required. Array of message objects with role and content.
`temperature`	number	Sampling temperature (0-2). Default: 1.
`max_tokens`	integer	Max tokens in the response.
`top_p`	number	Nucleus sampling threshold.
`stream`	boolean	Stream partial results via SSE. Default: false.
`stop`	string/array	Stop sequences.

Streaming

Set stream: true to receive partial results as server-sent events:

curl https://api.closedrouter.net/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-4o-mini",
    "stream": true,
    "messages": [{"role": "user", "content": "Count to 5"}]
  }'

data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","choices":[{"delta":{"content":"1,"},"index":0}]}
data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","choices":[{"delta":{"content":"2,"},"index":0}]}
data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","choices":[{"delta":{"content":"3,"},"index":0}]}
data: [DONE]

Using the OpenAI SDK

Set the base_url to ClosedRouter. Everything else stays the same.

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.closedrouter.net/v1",
    api_key="YOUR_API_KEY"
)

response = client.chat.completions.create(
    model="openai/gpt-4o-mini",
    messages=[{"role": "user", "content": "Hello!"}]
)

print(response.choices[0].message.content)

TypeScript / Node

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.closedrouter.net/v1",
  apiKey: "YOUR_API_KEY",
});

const response = await client.chat.completions.create({
  model: "openai/gpt-4o-mini",
  messages: [{ role: "user", content: "Hello!" }],
});

console.log(response.choices[0].message.content);

For AI agents and tools

ClosedRouter works with any tool that accepts an OpenAI-compatible endpoint:

OPENAI_BASE_URL=https://api.closedrouter.net/v1
OPENAI_API_KEY=YOUR_API_KEY

Errors

Status	Code	Meaning
401	`authentication_error`	Invalid or missing API key.
403	`forbidden`	Account does not have access to this model.
404	`model_not_found`	The requested model does not exist.
429	`rate_limit_exceeded`	Too many requests. Retry after the header.
500	`upstream_error`	The upstream provider returned an error.
502	`bad_gateway`	Could not reach the upstream provider.

Rate limits

During closed beta, rate limits are generous:

Limit	Value
Requests per minute	60
Tokens per minute	200,000
Max context length	Model-dependent

Rate limit headers are included in every response:

x-ratelimit-limit: 60
x-ratelimit-remaining: 58
x-ratelimit-reset: 30