You can now add your own OpenAI API key to your PerksPro+ account and unlock the ability to select different OpenAI models for each bot — like GPT-4 or GPT-4.0-MINI, or GPT-4.5-PREVIEW
This gives you:
-
More control over performance and speed
-
The ability to test and compare models per use case
-
Access to newer or more capable models (based on your OpenAI key)
๐ง How to Add Your OpenAI API Key
-
Log in to your PerksPro+ account
-
Click your profile name in the top-right corner
-
Select “Profile”
-
Scroll down to the OpenAI API Key field
-
Paste your API key here
โก๏ธ Need a key? Create one here
๐ค How to Select a Model for Your Bot
Once your API key is added:
-
Go to your Bots tab
-
Click on the bot you want to use and click Edit
-
In the Settings tab, scroll to Model Selection
-
Choose the model you want (e.g.,
gpt-4o,gpt-4o-mini, gpt-4.5-preview) -
Save your changes
Each bot can use a different model, so you can optimize for performance or cost depending on the bot’s job!
๐ณ Don’t Forget to Fund Your OpenAI Account!
To use your own OpenAI API key with PerksPro+, you’ll need to add funds to your OpenAI billing account. This ensures your bots stay active and can respond to users.
๐ How to Add Funds:
-
Visit your OpenAI Billing Settings
-
Add at least $5–$10 to start
-
You’re good to go — the usage cost will be deducted automatically based on the model and number of tokens used
๐ง Tip: One message is usually 50–100 tokens, so $10 gives you thousands of chats — especially with GPT-3.5 or 4o Mini!
๐ก FAQs
Do I need my own API key to use this feature?
Yes — model selection is only available if you’ve added your own OpenAI API key.
Will this cost more?
You’ll be billed directly by OpenAI based on your usage. Learn more about OpenAI pricing here.
Can I use Claude or Gemini instead of OpenAI?
Not yet — but we plan to expand support to other AI providers soon.
๐งช Best Use Cases
-
Use
GPT-4for your premium or high-complexity bots -
Use
GPT-3.5for lead gen or simple support flows to save tokens -
Test side-by-side to compare performance




