API Access

1. Create an API key

Sign in at anypath.dev, open Console -> API access, and create a key. Copy it immediately because the full secret is shown only once.

Keep the default intelligent route policy to route across eligible healthy shared capacity automatically. A route is one saved provider credential plus the models it serves; the key’s route policy decides whether AnyPath picks routes automatically (intelligent routing) or stays inside one provider (specified provider) — still not tied to one provider key. New to routes, providers, and multipliers? See Concepts.

Keep API keys in server-side secrets only. Do not expose them in browsers, mobile apps, public repositories, logs, or screenshots.

2. Pick the base URL

Protocol	Base URL	Main endpoints
OpenAI-compatible	https://api.anypath.dev/v1	/models, /chat/completions, /responses, /images/*
Anthropic Messages	https://api.anypath.dev	/v1/messages, /v1/messages/count_tokens
Gemini /v1beta	https://api.anypath.dev/v1beta	/models, /models/{model}:generateContent, /models/{model}:streamGenerateContent, /models/{model}:countTokens

3. Discover available models

curl https://api.anypath.dev/v1/models \
  -H "Authorization: Bearer sk-your-api-key"

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-api-key",
    base_url="https://api.anypath.dev/v1",
)
models = client.models.list()
for m in models.data:
    print(m.id)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk-your-api-key",
  baseURL: "https://api.anypath.dev/v1",
});
const models = await client.models.list();
models.data.forEach((m) => console.log(m.id));

The response only includes models available to that API key and route policy. Use this result as the source for model pickers and saved configuration.

4. Send a gateway request

curl https://api.anypath.dev/v1/chat/completions \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [
      {"role": "user", "content": "Summarize the attached support case in one paragraph."}
    ],
    "max_tokens": 300
  }'

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "Summarize the attached support case in one paragraph."}],
    max_tokens=300,
)
print(response.choices[0].message.content)

const response = await client.chat.completions.create({
  model: "gpt-4o-mini",
  messages: [{ role: "user", content: "Summarize the attached support case in one paragraph." }],
  max_tokens: 300,
});
console.log(response.choices[0].message.content);

Use POST /v1/responses, /v1/messages, or Gemini /v1beta only when the client needs that protocol shape.

Production checklist

• Call GET /v1/models before saving model configuration.
• Set explicit request timeouts and output limits.
• Retry only temporary failures such as 429, 502, 503, and 504; respect Retry-After when present.
• Check Console pricing, shared-route balance, and key usage limits before sending production traffic.
• Shared routes settle after successful requests.

Next

• Authentication: supported auth headers.
• API Overview: endpoint reference.
• Pricing and billing: how the price is set and what you are charged.
• Supplier guide: connect official capacity and earn USDT.

Quickstart covers the 5‑minute caller path (key → base URL → first request). The Supplier guide covers connecting provider credentials and earning from them.