OpenClaw Setup: API Key, Base URL, and Custom Models

OpenClaw connects to BetterToken through ~/.openclaw/openclaw.json. GPT provider models use openai-responses; other supported providers use openai-completions.

Key settings

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

Prerequisites

Install the latest OpenClaw
Create a BetterToken API Key
Copy a Model ID from the model plaza or key Setup dialog

Install

macOS / Linux
Windows

curl -fsSL https://openclaw.ai/install.sh | bash

iwr -useb https://openclaw.ai/install.ps1 | iex

Command-line setup

The BetterToken Automatic setup script writes the OpenClaw configuration. It requires Node.js. If you do not pass an API Key or Model ID as an argument, the script prompts you for them.

macOS / Linux
Windows

curl -fsSL "https://bettertoken.ai/install-openclaw-provider.sh" | bash

iwr "https://bettertoken.ai/install-openclaw-provider.ps1" -OutFile "$env:TEMP\install-openclaw-provider.ps1"; powershell -ExecutionPolicy Bypass -File "$env:TEMP\install-openclaw-provider.ps1"

Confirm that agents.defaults.model.primary uses the YOUR_MODEL_ID shown in Setup, then run the validation and Gateway restart commands below.

Manual setup

Configure OpenClaw

Edit ~/.openclaw/openclaw.json. Use only the example that matches the provider of your Model ID.

GPT: openai-responses
Non-GPT: openai-completions

{
  "models": {
    "mode": "merge",
    "providers": {
      "bettertoken": {
        "baseUrl": "https://bettertoken.ai/v1",
        "apiKey": "YOUR_API_KEY",
        "api": "openai-responses",
        "models": [
          {
            "id": "YOUR_MODEL_ID",
            "name": "YOUR_MODEL_ID"
          }
        ]
      }
    }
  },
  "agents": {
    "defaults": {
      "model": {
        "primary": "bettertoken/YOUR_MODEL_ID"
      }
    }
  }
}

{
  "models": {
    "mode": "merge",
    "providers": {
      "bettertoken": {
        "baseUrl": "https://bettertoken.ai/v1",
        "apiKey": "YOUR_API_KEY",
        "api": "openai-completions",
        "models": [
          {
            "id": "YOUR_MODEL_ID",
            "name": "YOUR_MODEL_ID"
          }
        ]
      }
    }
  },
  "agents": {
    "defaults": {
      "model": {
        "primary": "bettertoken/YOUR_MODEL_ID"
      }
    }
  }
}

Do not infer api from a model name. GPT uses openai-responses. Use openai-completions for a non-GPT provider only when the model plaza or Setup dialog confirms compatibility.

Verify the connection

Run:

openclaw config validate
openclaw gateway restart
openclaw models list
openclaw models status

The setup is active when validation passes, the Gateway restarts, and bettertoken/YOUR_MODEL_ID appears in the model list and status. If an existing session still uses an old model, start a new session and check again.

Switch models

Add the new model to models.providers.bettertoken.models, then change agents.defaults.model.primary to bettertoken/YOUR_MODEL_ID. Save, validate, and restart the Gateway.

Common errors

Error	Fix
`config validate` fails	Check JSON commas, quotes, and brackets.
`401`	Copy `apiKey` again and remove spaces.
`404` or protocol error	GPT must use `openai-responses`; non-GPT must use `openai-completions`. Do not append an endpoint to the Base URL.
Model missing	Confirm it exists in both the provider model list and `agents.defaults.model.primary`.
Existing session uses the old model	Run `openclaw gateway restart` and create a new session.

Advanced setup

Supported providers

Provider	Status
Claude	Not supported
GPT	Command-line + manual setup
Kimi	Manual setup
GLM	Manual setup

Statuses apply to the BetterToken setup documented on this page.

What the setup methods mean

Command-line + manual setup: use a generated command or follow the complete manual steps.
Manual setup: enter the API Key, Base URL, and Model.
Not supported: no verified direct setup is currently available.

Optional: manage the provider with CC Switch

To manage providers for several tools in one place, see OpenClaw setup in CC Switch.

Technical details

Responses and Chat Completions

openai-responses makes OpenClaw call /v1/responses for GPT provider models. openai-completions calls /v1/chat/completions for confirmed compatible non-GPT providers. Both use https://bettertoken.ai/v1 as baseUrl.

Getting started

Coding Agent setup

AI application platform setup

Advanced

OpenClaw Setup: API Key, Base URL, and Custom Models

Key settings

Prerequisites

Install

Command-line setup

Manual setup

Configure OpenClaw

Verify the connection

Switch models

Common errors

Advanced setup

Supported providers

Optional: manage the provider with CC Switch

Technical details

​Key settings

​Prerequisites

​Install

​Command-line setup

​Manual setup

​Configure OpenClaw

​Verify the connection

​Switch models

​Common errors

​Advanced setup

​Supported providers

​Related FAQ

​Optional: manage the provider with CC Switch

​Technical details

Key settings

Prerequisites

Install

Command-line setup

Manual setup

Configure OpenClaw

Verify the connection

Switch models

Common errors

Advanced setup

Supported providers

Related FAQ

Optional: manage the provider with CC Switch

Technical details