Is Routedly compatible with the OpenAI SDK?

Yes. Routedly speaks the OpenAI protocol. Set base_url to https://api.routedly.ai/v1 and keep the SDK you already use.

Per token, against a single balance. Each model has a published per-million-token rate for input, output and cache reads.

Which models can I call?

Models from OpenAI, Anthropic, Google, xAI, Alibaba, DeepSeek and others through one endpoint. GET /v1/models lists everything available to your key.

Quickstart

Fastest way to start using Routedly in any language or framework.

Routedly is a single drop-in endpoint that lets you call today’s best large-language models while keeping your existing code and development workflow intact. Point your HTTP requests at https://api.routedly.ai/v1/….

1 · Get an API key

Sign in to the dashboard.
Create a new Key → Copy the key. It is shown once, so store it before you close the dialog.

2 · Pick your language

Routedly speaks the OpenAI protocol, so every official OpenAI SDK works unchanged once its base URL points here.

bash

curl -X POST https://api.routedly.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $ROUTEDLY_API_KEY" \
  -d '{
    "model": "openai-gpt-4o",
    "messages": [
      {"role": "user", "content": "Hello, how are you?"}
    ]
  }'

3 · Going further

Streaming: pass stream: true on any request and Routedly proxies the event stream through unchanged. Streamed calls are billed on the tokens actually used, the same as any other call.
Monitoring: every call appears in the dashboard with its model, token counts and cost.

4 · FAQ

Is Routedly compatible with the OpenAI SDK?: Yes. Routedly speaks the OpenAI protocol. Set base_url to https://api.routedly.ai/v1 and keep the SDK you already use.
How do I authenticate with Routedly?: Every request takes a Bearer token in the Authorization header. Generate keys from your client dashboard.
How is usage billed?: Per token, against a single balance. Each model has a published per-million-token rate for input, output and cache reads.
Which models can I call?: Models from OpenAI, Anthropic, Google, xAI, Alibaba, DeepSeek and others through one endpoint. GET /v1/models lists everything available to your key.

Authentication

All requests require a Bearer token in the Authorization header. Generate keys from your client dashboard.

bash

curl https://api.routedly.ai/v1/models \
  -H "Authorization: Bearer sk-sr-cli-..."

Keys follow the format sk-sr-cli-<id>. Keys can be scoped with spend limits and disabled at any time from the dashboard.

Chat Completions

Identical schema to OpenAI's /v1/chat/completions — works as a drop-in replacement.

bash

curl https://api.routedly.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-sr-cli-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic-claude-sonnet-4-6",
    "messages": [
      {"role": "user", "content": "What is the capital of France?"}
    ],
    "stream": false
  }'

javascript

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.routedly.ai/v1",
  apiKey: process.env.ROUTEDLY_API_KEY,
});

const stream = await client.chat.completions.create({
  model: "openai-gpt-4o",
  messages: [{ role: "user", content: "Hi" }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content || "");
}

Models endpoint

Returns the full list of models available to your key. Use the id field directly as the model parameter in requests — e.g. anthropic-claude-sonnet-4-6, openai-gpt-4o, zhipu-glm-4-6. A model ID that is not in that list is refused with a 403 — the gateway bills against its own catalog, so an upstream provider’s native name is not a valid model value.

bash

curl https://api.routedly.ai/v1/models \
  -H "Authorization: Bearer sk-sr-cli-..."