Use isoboard with an OpenAI API key
You can use isoboard with your own OpenAI API key. Create a key in your OpenAI account, paste it into the OpenAI row on the Providers tab, and pick an image-capable model in a Profile. isoboard stores the key in the macOS Keychain, and OpenAI bills you directly for usage.
How do I connect OpenAI?
Section titled “How do I connect OpenAI?”- Sign in to the OpenAI API keys page and create a new secret key. Copy it when it is shown.
- In isoboard, open Settings and select the Providers tab.
- Expand OpenAI and paste the key into API key.
- Click outside the field. A status below it shows Ready or Configured. Configured means the key is saved and isoboard has not used it yet.
isoboard then creates two starting Profiles, OpenAI API - Efficient and OpenAI API - Accurate. Efficient favors faster, lower-cost captures. Accurate favors difficult captures and may take longer and cost more. You can edit or delete both. See Profiles.
Which OpenAI models work?
Section titled “Which OpenAI models work?”Any OpenAI model that accepts images as input. In the Profiles tab, choose OpenAI as the Provider, then search the model picker by name or type a model ID. Models that do not accept images fail at capture time. isoboard does not block them when you save.
Where does isoboard store my key?
Section titled “Where does isoboard store my key?”In the macOS Keychain, never in the configuration file. isoboard does not show the key again after you save it, and you replace it by pasting a new one. See where isoboard stores your data.
Who pays for the usage?
Section titled “Who pays for the usage?”You do. Each capture sends an image to OpenAI, and OpenAI charges your account under its own pricing. isoboard does not add a fee. The isoboard license is separate. See pricing and licensing.
What do OpenAI errors mean?
Section titled “What do OpenAI errors mean?”| Message in isoboard | Likely cause |
|---|---|
| the API key is invalid, or provider authentication is required | The key was revoked, mistyped or belongs to another account. Paste a new key. |
| provider access was denied | Your account or project cannot use that model. |
| provider inference is rate limited | You hit an OpenAI rate limit or ran out of credit. Wait, or check billing. |
| inference timed out | The request took longer than the Profile’s timeout. |
More fixes: provider errors.