Hi, I just started using the OpenAI API today following the quickstart. I have not even ran a successful query yet. My account says $0.00/$18.00 limit. It also says " You’ve used $0.00 out of the $18.00 total credit granted to you." Also, my activity tab is flat but I don’t know why I keep getting the below error:
openai.RateLimitError: Error code: 429 - {'error': {'message': 'You exceeded your current quota, please check your plan and billing details.', 'type': 'insufficient_quota', 'param': None, 'code': 'insufficient_quota'}}
How large is your prompt? Depending on that and what model you’re using you might not have a high enough tokens per minute (TPM) threshold to send the prompt.
It doesn’t make much sense to me. I am basically running the quickstart. I see it uses gpt-3.5-turbo and according to my account, I can see 40,000 TPM, 3 RPM, 200 RPD.
This is what I am running (from the docs):
from openai import OpenAI
client = OpenAI()
completion = client.chat.completions.create(
model="gpt-3.5-turbo",
messages=[
{"role": "system", "content": "You are a poetic assistant, skilled in explaining complex programming concepts with creative flair."},
{"role": "user", "content": "Compose a poem that explains the concept of recursion in programming."}
]
)
print(completion.choices[0].message)
I have signed up for ChatGPT and indirectly for OpenAI (because I am using the same email for both) many months back (surely more than 3). I am still using ChatGPT easily.
Solution 1: Generate an API key in a project or as a user key in billing; this is required to activate cel phone number verification,
Solution 2: Wait, the provisioning of an account and models takes some time,
Solution 3: Don’t trust the UI. You might see credits, but not be able to spend them until card processing is done by an outsourced provider.
Solution 4: Go to the chat playground, ensure the same organization and project is selected, see if you are blocked from even trying still.
Solution 5: Go to rate limits for the organization, see that models like GPT-4o have been turned on.
Solution 6: Generate another API key. List models by the models API endpoint. Make API calls with new key. Ensure it is an OPENAI_API_KEY system environment variable to be used by SDKs.
Solution 7: use “help” through the platform site and complain about this tedious and non-obvious problematic sign-up procedure that no other AI provider experiences.