How to Use TokenAir with WorkBuddy
A WorkBuddy job can trigger several model turns. Running suitable jobs through TokenAir can reduce that spend significantly, while sensitive or difficult work keeps a stronger model. WorkBuddy's Custom model flow accepts a URL, API key, and model name for OpenAI-compatible Chat Completions services.
Compatibility verdict
Direct setup
Direct setup is available through WorkBuddy's visual Custom model settings. Its standard mode appends and validates the /chat/completions path.
The tool documents a custom OpenAI-compatible endpoint path.
Before you start
- A current WorkBuddy version with Settings > Models > Custom available.
- A TokenAir API key and exact account-enabled model ID.
- A low-risk desktop task for the first test. WorkBuddy can operate on local applications and files.
Setup and compatibility steps
Step 1
Open the custom model settings
In WorkBuddy, open Settings, choose Models, add a model, and select Custom or Custom API.
Step 2
Enter the TokenAir values
Use https://api.tokenair.ai/v1 as the URL, paste the TokenAir API key into the local credential field, and enter one exact TokenAir model ID.
Step 3
Keep standard protocol mode on
Leave Custom protocol off for the first test so WorkBuddy uses the standard /chat/completions path. Custom protocol is for services with a non-standard full request URL.
Step 4
Run a low-risk task
Start with a task that cannot publish, delete, or send data. Confirm the output and model behavior before allowing broader desktop actions.
Values to enter in WorkBuddy
Provider: Custom
URL: https://api.tokenair.ai/v1
API Key: YOUR_TOKENAIR_API_KEY
Model: YOUR_TOKENAIR_MODEL_ID
Custom protocol: OffReplace placeholders with values from your own TokenAir account. Never commit an API key to source control.
How to verify it
- WorkBuddy saves the custom model and shows it in the conversation model picker.
- A text-only request completes without URL, key, or model validation errors.
- A representative desktop task finishes without repeated model loops.
- The selected model meets the task's privacy, tool-use, and quality requirements.
How to lower cost without hiding quality loss
TokenAir gives the client access to premium and lower-cost model choices through one OpenAI-compatible API. Savings depend on matching each task with a model that still passes review.
- Use a lower-cost model for repetitive document, extraction, or organization tasks with reviewable outputs.
- Keep risky computer actions confirmation-gated and use stronger models when an error would be expensive.
- Watch the TokenAir account for repeated calls because one desktop instruction can create multiple model turns.
Limits to know before production
- WorkBuddy stores the custom model credential locally. Protect the device and remove the key when it is no longer used.
- The external provider bills custom model usage. WorkBuddy may still charge credits for fixed-model features.
- Test tool calling and image input as separate cases. WorkBuddy may save a provider form even when the model lacks one of those capabilities.
FAQ
What URL should I enter for TokenAir in WorkBuddy?
Use https://api.tokenair.ai/v1 with standard protocol mode. WorkBuddy documents that it will validate and append the standard /chat/completions path.
Should I enable WorkBuddy Custom protocol?
Not for the normal TokenAir base URL. Custom protocol is intended for a full non-standard request URL and disables normal path completion.
Does a custom model remove all WorkBuddy charges?
Do not assume that. WorkBuddy states that external model usage is billed by the third party, while some fixed-model or product features may still use WorkBuddy credits.
Official sources and related TokenAir docs
Tool settings change between releases. We checked these notes on July 17, 2026. Review the linked tool docs again before a production rollout.
Try one workflow before changing the default.
Use the cost checkup to choose a bounded test, then compare cost per accepted result with your current route.