thinking and does not produce anything

[iamqiuxuan]

Been keeping thinking state for more than half an hour, this is the file window and log file, I tried local mode and pinecone mode, pinecone seems to need me to upgrade my account, what should I do

[iamqiuxuan]

I checked my openai account and ran about 30,000 tokens, but no data changes either locally, or on pinecone

[HostileRespite]

Same. It's been maddening.

[chenzx2]

i get the same problem,the code stop here

[HostileRespite]

i get the same problem,the code stop here !

I think the problem may be our security programs isolating the AI, but I can't tell. I've given the folder owner permission and can't see the program blocking it... It could be windows defender, but again, I can't find where it's blocking or quarantining the program.

[PedzacyKapec]

getting the same problem. Also once i hit enter im getting this:
File "E:\temp\autoGPT\Auto-GPT\scripts\main.py", line 461, in
main()
File "E:\temp\autoGPT\Auto-GPT\scripts\main.py", line 365, in main
assistant_reply = chat.chat_with_ai(
^^^^^^^^^^^^^^^^^^
File "E:\temp\autoGPT\Auto-GPT\scripts\chat.py", line 126, in chat_with_ai
assistant_reply = create_chat_completion(
^^^^^^^^^^^^^^^^^^^^^^^
File "E:\temp\autoGPT\Auto-GPT\scripts\llm_utils.py", line 50, in create_chat_completion
raise RuntimeError("Failed to get response after 5 retries")
RuntimeError: Failed to get response after 5 retries

---

EDIT: I think I solve the problem, at least in my case:
Ive just noticed my free API key expired on the 1st of April.
Thought I could use it till i use the whole £18. Didnt know it was time based.

[SamPink]

I get the same issue. It definitely needs a higher priority for finishing goals / tasks

[PedzacyKapec]

I think I solve the problem, at least in my case:
Ive just noticed my free API key expired on the 1st of April.
Thought I could use it till i use the whole £18. Didnt know it was time based.

[tennessine]

In my case, it was network problem, when I use aliyun, it stucked, when use American server, it worked fine.

[JuroOravec]

@iamqiuxuan try to run the script with --debug flag.

@PedzacyKapec
Sharing for visibility - I was dealing with other issue (invalid OpenAI API key because the account wasn't converted to "paid account", see #1189). But it looks like there's the repeating pattern that when we get

Failed to get response after 5 retries

It's likely an issue with the OpenAI API key.

[egohug]

I use the free GPTapi and keep thinking about it

[Hashan-Peiris]

I get "Warning: The file 'auto-gpt.json' does not exist. Local memory would not be saved to a file.
Debug Mode: ENABLED
Welcome back! Would you like me to return to being CoPilotGPT?
Continue with the last settings?
Name: CoPilotGPT
Role: ###################
Goals: [####################]
Continue (y/n): y
Using memory of type: LocalCache
Token limit: 4000
Memory Stats: (0, (0, 1536))
Token limit: 4000
Send Token Count: 914
Tokens remaining for response: 3086
------------ CONTEXT SENT TO AI ---------------
System: The current time and date is Fri Apr 14 XXX 2023

System: This reminds you of these events from your past:

User: Determine which next command to use, and respond using the format specified above:

----------- END OF CONTEXT ----------------
Error: API Rate Limit Reached. Waiting 20 seconds...
Error: API Rate Limit Reached. Waiting 20 seconds..."

[frozenranger]

Same issue :(

[Qoyyuum]

1. upgrade to a paid account: https://platform.openai.com/account/billing/overview
2. ChatGPT+ != paid account.
3. add a payment method: https://platform.openai.com/account/billing/payment-methods (optional) set my usage limits: https://platform.openai.com/account/billing/limits
4. create a new API key: https://platform.openai.com/account/api-keys (old API key did not work, i guess it was not update with my payment information automatically)
5. replace the value for the autogpt/.env files's OPENAI_API_KEY.

Closing issue.

[anonhostpi]

I think this issue may have been closed prematurely. While most of the comments were about keys/rate-limits the original post was about api.openai.com/v1/embeddings being unreachable.

Not going to reopen, as I believe #3977 has the same problem