SideNote ProUse an OpenAI API key with SideNote Pro
OpenAI is the quickest concrete BYOK setup, which is why the application's own guide uses it as the example. This is the same walkthrough with more of the reasoning behind each field.
Creating an API key
- Sign in to the OpenAI developer platformThis is the API platform, which is a separate product from the consumer ChatGPT subscription. A ChatGPT Plus subscription does not include API credit.
- Add a payment method and set a spending limitDo the limit now rather than later. A hard cap in the provider dashboard is the only limit that is authoritative.
- Create a secret keyCopy it immediately. Providers typically show a secret key once and never again.
Treat the key like a password with a credit card attached, because functionally it is one. Do not paste it into a chat, a screenshot, a support ticket or a repository. If a key is ever exposed, revoke it in the provider dashboard and issue a new one; that costs nothing.
Configuring the profile
In SideNote Pro, open Settings, then AI providers, and add a profile from the OpenAI preset. The fields it needs are these.
| Field | Value |
|---|---|
| Provider name | OpenAI |
| Base URL | https://api.openai.com/v1 |
| Model | gpt-5-mini |
| API key | Your own secret key |
| Reasoning effort | Provider default |
gpt-5-mini is what a new OpenAI profile starts on, and it is a sensible everyday default rather than a ceiling. Once the connection test works, model discovery gives you the full list your account can reach, and choosing the right AI model covers how to pick between them.
In examples on this site, a key is written as sk-example-not-a-real-key. That is a placeholder. Real keys are issued only by your provider.
Test the connection before anything else
Run the profile's connection test. It calls GET https://api.openai.com/v1/models with your key as a bearer token. If that succeeds, the base URL is right, the key is valid, and the discovered model IDs populate the searchable model picker.
| Symptom | Likely cause |
|---|---|
| Credentials rejected | Key mistyped, revoked, or from a different account |
| Endpoint not found | Base URL missing the /v1 segment, or a trailing path added |
| Endpoint unreachable | No network, a proxy in the way, or a firewall rule |
| Rate limited | Quota or spending limit reached on the provider side |
| Model rejected on send but test passes | The key is fine; that specific model is not available to your account |
What SideNote Pro actually sends
You never construct an HTTP request yourself. But knowing the shape helps enormously when you are debugging a compatible provider later, and it tells you which documentation applies.
GET https://api.openai.com/v1/models
Authorization: Bearer sk-example-not-a-real-keyPOST https://api.openai.com/v1/chat/completions
Authorization: Bearer sk-example-not-a-real-key
Content-Type: application/json
{
"model": "gpt-5-mini",
"stream": true,
"messages": [ ... ]
}The response comes back as Server-Sent Events, data: records terminated by [DONE], which is what lets text appear as it is generated rather than arriving in one block.
SideNote Pro uses Chat Completions, not the newer OpenAI Responses API. This trips people up because a lot of current documentation and blog writing has moved to Responses without saying which generation it means. If you are diagnosing a compatibility problem, Chat Completions is the contract that applies.
Also not currently used: tool calling, built-in web search, audio, video and image generation. SideNote Pro sends text and images and streams text back.
Understanding what you will be charged
API billing is per token, priced separately for input and output, and it varies by model. We are not going to print a price table here, because provider prices change and a stale number on a website is worse than no number. Check your provider's own pricing page.
What you can do inside SideNote Pro is record those prices per profile and model in USD per million tokens, which turns the local ledger into meaningful estimates. See usage and cost tracking, and reducing your AI API costs for the habits that matter most.