Hi, Can you take a look at his table please. I'm trying to run a claygent websearch on 15k colomns but it's been stuck a around 2.3k for 5 days now. I increased my limits all the way to Tier 4 with open AI so I should be able to process all of these fairly quickly but nothing happens when I try, the rols just are stuck in queued
For Claygent to function properly, you need a minimum of 450,000 TPM (tokens per minute) from your OpenAI API.To address the processing delay, try these steps:1. Break your 15k columns into smaller batches 2. Verify your search queries are specific and well-defined 3. Double-check your prompts and search parametersEven though you've upgraded to Tier 4, please verify your current TPM limit at: https://platform.openai.com/account/limits to ensure you have the required 450,000 TPM for Claygent operations.
Looks like you're still looking for help. We're escalating this over to our support team and someone should be in touch soon. In the meantime, feel free to include any other information that may be helpful in solving your issue!
The colomn is Case Study Company
I tried running just 1k then 200 rows but still only get a handful processed then they get stuck
The first day we processed 2k or so and that was with Tier 2 so I don't see any reason we can't process 1000s more with our new Tier 4 limits
Hi Devon, thanks for reaching out. Taking a look at this now. Can you share a screenshot of the lower section of the page at this URLso I can take a look at your usage tier here?
sure, here it is I'm tier 4
The GPT-4o-mini model might not be ideal for this task. Switching the Claygent task to run on the GPT-4o model could be a better option to avoid these timeout issues.
Hey there - just wanted to check in here to see if you needed anything else! Feel free to reply back here if you do.