# Use BigModel.io with Cherry Studio

> Add BigModel.io to Cherry Studio as a custom OpenAI-type provider with API address https://bigmodel.io, fetch GLM, Kimi and DeepSeek models, and check the connection.

Web version: https://bigmodel.io/docs/cherry-studio · Updated 2026-10-18

Cherry Studio is a desktop chat client with custom model providers. Add BigModel.io as an **OpenAI**-type provider and it calls Chat Completions, which every text group serves.

| Field in Cherry Studio | Value |
|---|---|
| Provider type | OpenAI |
| API Key (API 密钥) | your BigModel.io API key (`sk-...`) |
| API address (API 地址) | `https://bigmodel.io` (Cherry Studio appends `/v1/chat/completions`) |
| Models | model IDs from the [model catalog](https://bigmodel.io/models), e.g. `glm-5.3` |

Unlike most tools, the OpenAI-type address here is the root **without** `/v1`. If you enter `https://bigmodel.io/v1`, requests go to `/v1/v1/chat/completions`.

## Add the provider

1. Open **Settings → Model Services (模型服务)**.
2. Click **+ Add (添加)**, enter a name such as `BigModel.io`, and choose the provider type **OpenAI**.
3. Paste your key from [API Keys](https://bigmodel.io/keys) into **API Key**. Several keys can be entered separated by commas.
4. Enter **API address** `https://bigmodel.io`.
5. Add models: click **Fetch model list (获取模型列表)** and add the ones you want, or click **+** and enter an exact model ID such as `glm-5.3`. Only models you add appear in the chat model picker.
6. Turn the provider on.

To send a full URL verbatim, end the address with `#`, for example `https://bigmodel.io/v1/chat/completions#`. You normally do not need this.

## Verify the setup

Click **Check (检查)** next to the API key, pick one of your added models, and confirm it succeeds. Then start a chat with that model and check that the request appears in [Usage](https://bigmodel.io/usage).

## Troubleshooting

**Check fails with 404.** The address has `/v1` or a full path. Use `https://bigmodel.io` for the OpenAI type.

**The model list is empty or fetching fails with 403 `INSUFFICIENT_BALANCE`.** Listing models requires a balance above $0. Top up in [Billing](https://bigmodel.io/purchase), or add models manually with **+**.

**503 `api_error`: `No available accounts: no available accounts supporting model: ...`.** The model you added is misspelled or not served by your key's group. Copy the ID from the model page.

**401 `INVALID_API_KEY`.** The key has spaces or was deleted. Copy it again from [API Keys](https://bigmodel.io/keys).

**A model does not show in the chat picker.** Only added models appear. Add it under the provider's model list.

**429 `API_KEY_QUOTA_EXHAUSTED` or `rate_limit_exceeded`.** The key's quota or a 5h/1d/7d spend window is used up. Raise or reset it in [API Keys](https://bigmodel.io/keys), or wait for the window to reset.

## Related

- [LobeChat](https://bigmodel.io/docs/lobechat) for a self-hosted web chat client.
- [CC Switch](https://bigmodel.io/docs/cc-switch) for switching coding tools between providers.
- [Authentication](https://bigmodel.io/docs/authentication) for how keys and groups work.