gpt-engineer
gpt-engineer copied to clipboard
gpt-engineer-org
ONLY works with GPT-4 access??
Where are you declaring the Model? I don't see it anywhere, and I am still waiting for access to the GPT-4 API. From what I understand of late the 3.5 turbo API is pretty much just as good anyways no? At any rate, I'd really like to try this out. How can I change the model to 3.5-turbo?
ok I've got it working, the issue was I had to manually paste the main_prompt file into the newly created project folder. but its unable to recognize and correct itself when it makes an error. it tried to run main.py when it hadn't yet created a main.py file, and then aborted rather than recognizing that it needed to create the file.
I tried to test using 3.5 turbo but did not work. Have you figured out working with 3.5?
Specifying model it can work, I guess.
!gpt-engineer projects/example --model 'gpt-3.5-turbo-0613'
I tried to test using 3.5 turbo but did not work. Have you figured out working with 3.5?
what do you mean it did not work? as in did not run at all? or kept making mistakes? i did get it to run, but it keeps making mistakes for me. i'll try specifying the model and see it that helps
I have changed the gpt4 to gpt-3.5-turbo-0613 but I keep getting openai.error.RateLimitError . Any ideas where this might be coming from. Trying to run the example snake
@LindroosKE "A RateLimitError indicates that you have hit your assigned rate limit. This means that you have sent too many tokens or requests in a given period of time, and our services have temporarily blocked you from sending more."
it doesn't seem to be working properly but i've given it a complex use case, I'm trying to have it build a social media management tool. it has created a generic 'social-, media-api' file, and says I'm supposed to input my 'social media api' before running it again. maybe i need to be more specific in my instructions. it also keeps trying to run a 'main.py' file that does not exist, and then it stops running. initially i thought it was maybe adding to its work every time it started again but i guess not because each time i run it again it creates a new virtual environment folder to work in. which means it keeps starting over.
i wonder if it will work better once i have a gpt-4 api
i wish Anton Osika would provide more documentation for how to run it properly too
On Sun, Jun 18, 2023, 9:39 PM ma-raza @.***> wrote:
I tried to test using 3.5 turbo but did not work. Have you figured out working with 3.5?
— Reply to this email directly, view it on GitHub https://github.com/AntonOsika/gpt-engineer/issues/177#issuecomment-1596482195, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHRLYSZLSZKDFWH7OZPA6MLXL7JYJANCNFSM6AAAAAAZLGNMZQ . You are receiving this because you authored the thread.Message ID: @.***>
If you wish to reopen the issue please do following the new issue template.
