APIMODELSAPIMODELS

How to connect Cherry Studio to apimodels.app

TL;DR In Cherry Studio open Settings, Model Provider, Add Provider. Name it, paste your apimodels.app key, and enter https://api.apimodels.app/v1 as the OpenAI base URL. Turn the provider on, click Sync models, and add the language models you want, such as claude-opus-5-5, qwen4-max or gpt-6-sol. Use Model Check to confirm the key works.

Cherry Studio is an open-source desktop client for chatting with language models. It lets you add your own provider by entering an API address and a key, which is how you use apimodels.app in it: one key, and every model on that key appears in one provider.

The labels below were read from the Cherry Studio source and language files on 17 November 2026 (version 2.2.2). Older 1.x builds use a different dialog.

Add the provider

Go to Settings, then Model Provider, and click Add Provider at the bottom of the list. The Add Custom Provider dialog opens.

FieldWhat to enter
Provider NameAny name, for example apimodels
API KeyYour apimodels.app key
Endpoint settings: OpenAIhttps://api.apimodels.app/v1
Endpoint settings: Anthropic (optional)https://api.apimodels.app

How Cherry Studio builds the request address

Under each address field Cherry Studio shows a "Request path" preview, so you can see the final address before saving. For the OpenAI field it should read https://api.apimodels.app/v1/chat/completions, and for the Anthropic field https://api.apimodels.app/v1/messages.

Cherry Studio adds /v1 only when the address has no version in it, so https://api.apimodels.app and https://api.apimodels.app/v1 give the same result in the OpenAI field. Ending the address with # turns the automatic version off; you do not need that here.

You need at least one text endpoint. The OpenAI one is enough for every language model on your key, Claude included. Fill the Anthropic one only if you want Claude models to go through the Anthropic message format.

Turn it on and add models

After you click Add, open the new provider and turn on the switch at the top right. Then click Sync models next to the Models heading. Cherry Studio fetches the list from our model-list endpoint.

The list holds every model on your key, including image, video and audio models, which will not work in a chat window. Add only the language models you plan to use. A fetched model shows up in the chat model picker only after you add it to the provider's list.

To add one by hand, click the plus sign and enter the Model ID, for example claude-opus-5-5.

Check the connection

Click Model Check to the right of the API Key field. It sends a real request with your key to a model in your list, so it uses a small amount of balance. A pass means the address, key and model all work.

Limits and things that go wrong

A model answers but cannot see images or use tools. Open the model's edit panel and tick the matching items under Model Capabilities (Vision, Tool, Reasoning). Cherry Studio does not know the capabilities of models from a custom provider. Ticking a box only enables the feature in the interface; the model itself must support it.

Thinking models start slowly. qwen4-max finishes thinking before it writes, so the first words can take tens of seconds.

Image and video models are not for the chat window. Use them through the API. See the model pages for the request format.

Several keys. Cherry Studio accepts several keys separated by English commas and rotates between them.

Sources

Labels and address rules were read from the Cherry Studio repository (main branch, version 2.2.2) on 17 November 2026: the custom-provider dialog code, the API address formatter, the model-list fetcher and the language files. We did not test every Cherry Studio version.

Quick start

cURL

# Check the address and key outside Cherry Studio
curl https://api.apimodels.app/v1/chat/completions \
  -H "Authorization: Bearer $APIMODELS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-opus-5-5",
    "messages": [{"role": "user", "content": "Say hello in one word."}]
  }'

FAQ

What API address do I enter in Cherry Studio?

Enter https://api.apimodels.app/v1 in the OpenAI field. The preview under the field should read https://api.apimodels.app/v1/chat/completions.

Why are my models missing from the chat model picker?

Two things to check: the provider switch at the top right is on, and the model has been added to the provider's model list after Sync models.

Can I use Claude in Cherry Studio through apimodels.app?

Yes. Add claude-opus-5-5 or claude-sonnet-5-5 to the model list. They work through the OpenAI address; the Anthropic address is optional.

Does Model Check cost anything?

Yes, a very small amount. It sends one real request to the model you select.

Related

등재된 곳
APIMODELS on Product HuntAPIMODELS featured on ToolifyAPIMODELS featured on There's An AI For That