Authentication
Each provider has a different auth method. CCR + OpenRouter works out of the box; Claude and Codex require connecting your existing subscription.CCR + OpenRouter
No setup required. CCR is the default provider, managed entirely by Whim. API usage is billed as part of your CU allocation.Claude Subscription
Get your OAuth token
- The settings panel shows when your token was created and last used
- Paste a new token to replace the existing one
- Clear the field and save to disconnect
Codex Subscription
Enter the device code
Whim displays a device code and a link to the OpenAI verification page. Open the link in your browser.
Device codes expire quickly. If it expires before you finish, click Sign in to ChatGPT again.
Connection status
| Indicator | Meaning |
|---|---|
| Green dot — Connected | Authenticated and ready |
| Grey dot — Not connected | Needs setup |
| Spinner — Checking | Verifying connection |
Workspace and user defaults
Workspace defaults (admin)
Go to Settings > Workspace > Defaults to configure for all members:- Default provider and model
- Permission mode
- Fast mode
- Reasoning effort (Codex)
User defaults
Go to Settings > My Defaults to override workspace defaults for your tasks only. User defaults take precedence over workspace defaults. Per-task settings override both.Provider-specific settings
| Setting | CCR + OpenRouter | Claude | Codex |
|---|---|---|---|
| Model | Yes | Yes | Yes |
| Permission mode | Yes | Yes | Yes |
| Fast mode | No | Yes | Yes (GPT 5.4 only) |
| Reasoning effort | No | No | Yes |
Fast mode
Fast mode tells the provider to prioritize response speed over reasoning depth. Useful for straightforward tasks — simple edits, boilerplate, quick fixes.| Provider | Effect | Supported models |
|---|---|---|
| Claude | Passes --effort low to the CLI | All Claude models |
| Codex | Sets service_tier=fast | GPT 5.4 only |
| CCR | Not available | — |
Fast mode for Claude uses the same model — it reduces reasoning effort, not model size.
- Per task — toggle when creating a task
- User default — Settings > My Defaults
- Workspace default — Settings > Workspace > Defaults
Codex has fast mode enabled by default for GPT 5.4. Claude has fast mode disabled by default.
Reasoning effort (Codex only)
Controls how much compute the Codex model spends thinking before responding. Higher effort = more thorough analysis but slower output.| Level | Best for | Speed |
|---|---|---|
| Low | Simple, well-defined tasks | Fastest |
| Medium | Moderate tasks, some reasoning helpful | Fast |
| High | Complex tasks benefiting from deeper analysis | Moderate |
| xhigh | Difficult problems — complex debugging, architecture, multi-file refactors | Slowest |
Speed vs quality reference
| Configuration | Speed | Quality | CU impact |
|---|---|---|---|
| Claude Opus 4.6 | Slower | Highest | More container time |
| Claude Opus 4.6 + fast mode | Faster | High | Less container time |
| Claude Sonnet 4.5 | Fast | High | Moderate |
| Claude Sonnet 4.5 + fast mode | Fastest (Claude) | Good | Least container time |
| Claude Haiku 4.5 | Very fast | Good | Low |
| Codex GPT 5.4 + xhigh reasoning | Slower | Highest (GPT) | More container time |
| Codex GPT 5.4 + low reasoning | Fastest (Codex) | Lower | Least container time |
Troubleshooting
Claude token shows 'Not connected' after saving
Claude token shows 'Not connected' after saving
Your OAuth token may have expired. Run
claude setup-token locally to generate a fresh token and paste it in Whim.Codex device code expired
Codex device code expired
Device codes are time-limited. Click Sign in to ChatGPT again to generate a new one.
Tasks fail immediately after switching providers
Tasks fail immediately after switching providers
Verify the provider is connected (green dot) before setting it as your default. Existing tasks keep the provider they were started with.
My user defaults aren't taking effect
My user defaults aren't taking effect
User defaults only apply to new tasks. Check Settings > My Defaults (not workspace defaults) and verify the provider is connected.