> ## Documentation Index
> Fetch the complete documentation index at: https://docs.bettertoken.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# GLM API in OpenCode: API Key, Base URL, and Model Setup

> Connect a GLM model to OpenCode through BetterToken. Set an API Key, Base URL, and the exact GLM Model ID with /connect or opencode.json.

You can connect GLM to OpenCode through BetterToken as an OpenAI-compatible provider. Enter your BetterToken API Key, the Base URL `https://bettertoken.ai/v1`, and the exact GLM Model ID from the Setup dialog.

## Key settings

| Field    | Value                       |
| -------- | --------------------------- |
| API Key  | BetterToken API Key         |
| Base URL | `https://bettertoken.ai/v1` |
| Model    | `YOUR_GLM_MODEL_ID`         |

## Prerequisites

* Install [OpenCode](https://opencode.ai/).
* <a href={"https://bettertoken.ai/register"}>Create a BetterToken API Key</a>.
* In the **Setup** dialog, select OpenCode and the **GLM** provider, then copy the available full Model ID.

## Configure with /connect

Start OpenCode and run:

```text theme={null}
/connect
```

Choose **Other**, enter `bettertoken` as the provider id, and paste your BetterToken API Key. Then select `bettertoken/YOUR_GLM_MODEL_ID` and restart OpenCode.

## Manual configuration

Open `~/.config/opencode/opencode.json` or a project-level `opencode.json`:

```json theme={null}
{
  "$schema": "https://opencode.ai/config.json",
  "model": "bettertoken/YOUR_GLM_MODEL_ID",
  "provider": {
    "bettertoken": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "BetterToken",
      "options": {
        "baseURL": "https://bettertoken.ai/v1",
        "apiKey": "YOUR_API_KEY"
      },
      "models": {
        "YOUR_GLM_MODEL_ID": {
          "name": "YOUR_GLM_MODEL_ID"
        }
      }
    }
  }
}
```

Replace `YOUR_API_KEY` and `YOUR_GLM_MODEL_ID` with values from the Setup dialog. Do not reuse a Model ID from an old article; use one currently available to your API Key.

## Verify the connection

Select `bettertoken/YOUR_GLM_MODEL_ID` in OpenCode and send a short prompt. A response without an API error confirms the setup. Fully restart OpenCode after editing JSON.

## Common errors

| Error                        | What to check                                                                             |
| ---------------------------- | ----------------------------------------------------------------------------------------- |
| `401`                        | Check the BetterToken API Key saved through `/connect` or in `apiKey`.                    |
| `404`                        | Keep `baseURL` as `https://bettertoken.ai/v1`, without `/chat/completions`.               |
| Model not found              | Copy the full GLM Model ID again from the Setup dialog and use it in both JSON locations. |
| Configuration is not applied | Check the project-level `opencode.json`, which can override the global file.              |

## Related guides

* [OpenCode setup](/en/ai-tools/opencode)
* [Kimi K3 in OpenCode](/en/faq/opencode/kimi-k3-api)
* [Claude Code setup](/en/ai-tools/claude-code)
* [Choose a Model ID](/en/faq/model-calling/model-selection-guide)

## Technical details

<Accordion title="Base URL and Model ID">
  OpenCode uses OpenAI-compatible Chat Completions. Set only `https://bettertoken.ai/v1` as the Base URL because OpenCode appends the request path. The available GLM Model ID depends on the selected provider and your API Key.
</Accordion>