Why did you throttle the capability for openai models to surf the web with our API key? I just signed up again after a couple month and now this feature is gone?
OpenAI's API itself cannot access live data from the internet - it can only reference data it was trained on. For web research in Clay, you need to use Claygent, our AI-powered web researcher that pulls information from websites in real-time. Claygent requires an OpenAI API key with Tier 2 or higher access (≥450, 000 TPM). If your API key doesn't meet these requirements, the web research feature won't be available.
I have a Clay account connected to my work email. I'm trying to reset my password but I'm locked out + I never receive the email to reset my password (it's not in my spam folder, the email is correct, etc).
Hey Dylnn, thanks for reaching out. You have Google Auth login enabled for your Clay account. Let me know if you'd like to change it to password login.
this wasnt the original question. I asked why you guys throttles the ability for my own gpt models to access the internet
Hey, We don't throttle your GPT models' internet access. For web research with your own OpenAI API key, you need to use Claygent with "Web Research" mode and your API
However, the requirement is having Tier 2 token limits on your OpenAI account (usually done by adding prepaid credits in your OpenAI account). There's a need for the higher limits because web research is complex and without them, requests fail and you'd still get charged.
This isn't throttling - it's ensuring the web research actually works reliably with your API key.
Let me know if you have more questions.