Suanlimao
← Back to all tutorials

Connect your tools

Easydict

The macOS dictionary and translator: add a “Custom OpenAI” service.

About 4 minFor:Mac users who translate with Easydict

Before you start

  • Easydict is installed (macOS)
  • A key and a model name

Step 1 · STEP 0x01

Add a Custom OpenAI service

Settings → Services
1
Add service
2

Service type

Custom OpenAI
Diagram only — the real screen in your version may differ

Open the Services page in Easydict settings and add a “Custom OpenAI” service. The docs describe it as the way to “configure an OpenAI-compatible endpoint”, and you can add several instances — one per model if you like.

Create a key →

Step 2 · STEP 0x02

Enter the key, endpoint and model

Custom OpenAI → service settings
1

API Key

<YOUR_API_KEY>
2

Endpoint

https://api.suanlimao.shop/v1/chat/completions
3

Model

<MODEL_NAME>
4

Enable the service

On
Diagram only — the real screen in your version may differ
  • The endpoint is the full request URL https://api.suanlimao.shop/v1/chat/completions: Easydict requests exactly that URL and does not append /v1/chat/completions.
  • The model is the name copied from the plaza.
The docs only mention entering your own API key for an OpenAI-compatible endpoint and do not list field labels; the labels above follow the current Easydict UI.
Don't guess the model name: open the model plaza, click “Copy model id” on a model card and paste it. A wrong model name is the most common error.

Open the model plaza →

Step 3 · STEP 0x03

Look up a word

Easydict query window
1

Input

serendipity
2

Service in the results

Custom OpenAI
Diagram only — the real screen in your version may differ
  • Look up a word; a translation in the Custom OpenAI row means it works.

✓ How to tell it worked

  • The Custom OpenAI row shows a translation and the Usage page shows the request.

FAQ

“Endpoint is invalid”?

The endpoint must be a complete https URL ending in /chat/completions.

“API key is empty”?

This service instance has no key, or you entered it in another instance.

HTTP 404?

The endpoint stops at /v1 — add /chat/completions — or the model name is wrong.

Related tutorials

Still stuck?

Read the troubleshooting tutorials first. If that does not help, open a ticket with the request_id, time, model and the exact error (never paste your key).