I asked just one line but it always return rate_limit_exceeded

API
1

I am testing with api model GPT-4o.
I am now at tier1.
I asked just one line but it always return rate_limit_exceeded error.
below is my logging.

asked : I'm so frustrated. What a rate_limit_exceeded error for a one-line question!
........................................................
openai assistant error : {"code"=>"rate_limit_exceeded", "message"=>"Request too large for gpt-4o in organization org-BlaBla on tokens per min (TPM): Limit 30000, Requested 30205. The input or output tokens must be reduced in order to run successfully. Visit https://platform.openai.com/account/rate-limits to learn more."}

Yesterday I tested a little and I got the payment receipt(5 dollars) .
Today, I started testing again but I always got this error.

Is this because my account is at Tier1?
If it is right, How can I move to Tier2?

2

1000020001
1000020001856×248 29.9 KB

3

Thanks for your answer.

I understood it is counted per chat so the error was weird to me.
I asked just one line of question with api.

So does this mean that tokens are accumulated and counted on a per-thread basis?

4

Tier 1 has an abysmal 30k per minute, 30k per model call API token limit to GPT-4o.

Assistants will easily overflow this with a search or two on documents added to a thread.

The message is clear: Prepay $50+ if you don’t want your account emptied with errors by assistants.

Or a lesson to read the API documentation for chat completions to use that and manage your input.

5

Thank you very much!!
I topped up my account with 50 dollars more, and it works!

6

when i hit api it shows this error again and again we get an error

Run status: failed
Assistant run failed.
2024-10-07 10:48:11 error: Error: Assistant run failed
at checkStatus (C:\theriome_today (1)\src\helpers\custom\openaiUtils.js:77:11)
at processTicksAndRejections (node:internal/process/task_queues:96:5)
at async Timeout. (C:\theriome_today (1)\src\helpers\custom\openaiUtils.js:111:28)

how to solve this and i dont able to find the main cause of assistant run failed?