Home
Events
Groups
Members

00 Clay Announcements
01 Community General
01 Props and Wins
01 Watercooler
01 Welcome and Intros
02 Bugs
02 Starting with Clay
02 Support
03 Feedback & Requests
03 Tutorials & Inspiration
04 FT Job Opps
Gtm Career Advice
Member Spotlight

Community Guidelines

Powered by Tightknit

How to Avoid Throttling When Using OpenAI API with Claygent

·

·

I have credits in open ai but I have tier 1 access to their api, and I keep getting this message "Could not finish the message because max_tokens or model output limit was reached. Please try again with higher max_tokens." when trying to run claygent. How can I use my api key without getting throttled?

How to Avoid Throttling When Using OpenAI API with Claygent | Clay

6 comments

· Sorted by Oldest

Joel Martinez
·
·
chat gpt 5 mini
Joel Martinez
·
·
https://app.clay.com/workspaces/688665/workbooks/wb_0t2diei4wGHBopSasG9/tables/t_0t2cdsu9KNFjYqsVGMJ/views/gv_0t2cdsucQQwj7mkczKk
Joel Martinez
·
·
I ended up using a different model, but chat gpt has not been working for me
Joel Martinez
·
·
do you know of a resource to understand which model to select in the context of output? and The original prompt I used was very short, but after using clay's native AI generator the complexity of prompt expanded significantly. Is there a best practice / resource for prompting that I can use to avoid this?
Yash Mehta (Clay)
·
·
So while the prompt may be short, the tasks that the model is doing in the background may be complex.
In general you could compare model strengths based on the task you want them to perform by clicking "Compare models" in your AI column setup (see image below).
Also for prompting best practices you can check this resource out - https://www.clay.com/university/lesson/prompt-engineering-crash-course-clay-101
Let me know if this helps!
Joel Martinez
·
·
Thanks