Skip to Content
Quickstart

Quickstart

Use this guide to create an Accly API key and send your first request through Accly Gateway.

The primary quickstart path is the OpenAI-compatible chat completions endpoint. For this endpoint, create an openai or universal key.

1. Create an account

Open the Accly app and sign up or sign in:

https://chat.accly.net 

The Accly app is where you create API keys, view model access, monitor usage, and manage billing.

2. Choose a plan with API access

Free does not support API keys.

To use the Gateway API, choose a paid plan:

PlanAPI accessModel tiers
StarterYesBasic, Advanced
ProYesBasic, Advanced, Thinking, Beta
MaxYesBasic, Advanced, Thinking, Beta

3. Create an API key

In the Accly app:

  1. Open settings.
  2. Go to API Keys.
  3. Create a key.
  4. Choose a key type for the endpoint family you plan to use.
  5. Copy the key immediately.

The full key is shown once. Store it in a server-side environment variable or secret manager.

For /v1/chat/completions, use an openai or universal key type. Provider-specific endpoint families require matching key types: use anthropic for Anthropic-style requests and google for Google/Gemini-style requests.

4. Choose a supported model ID

Choose a model ID from the Accly app model picker or account model list. Use that model ID in the model field of your API request.

Model availability and capabilities may vary by plan. If your plan cannot use the requested model tier, the gateway returns model_not_allowed.

5. Send a request to /v1/chat/completions

Replace sk-your-accly-api-key and gpt-5.5 with your own key and an available model ID.

First request
curl https://api.accly.net/v1/chat/completions \
  -H "Authorization: Bearer sk-your-accly-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.5",
    "messages": [
      {
        "role": "user",
        "content": "Write a three sentence product update."
      }
    ],
    "stream": false
  }'

6. Monitor usage

Use the Accly app to review plan, API keys, request-unit usage, rate limits, model access, and subscription status.

API usage can be limited by:

  • Daily request units
  • Request-count guard
  • RPM rate limit
  • Model tier access
  • Subscription or account status

One API call does not always equal one request unit. Heavier models, large context, long outputs, and agentic coding sessions can consume more units. Accly does not silently replace the model you choose with an unrelated cheaper model. See Pricing & Usage.

Next: connect an agent

After your first Gateway request works, connect an agent or coding tool with the same Accly key. Start with Codex Setup if you want Codex to use Accly Gateway with an OpenAI-type Accly API key.