Before getting on the codex waitlist I was getting an “engine not found” error when I tried to make an api call for it. After getting on the waitlist I am getting the same error. I tried spelling it “code-davinci-002” and even “code-davinci:002” but both yielded the same result.
How can I access the codex api? What do I need to do?
*NOTE: this is resolved as of 5/11/22
sps
May 11, 2022, 9:06pm
2
Hi @scottlydon18
Welcome to the OpenAI community.
Can you share more about the API call you made? Code? Endpoint? Response?
My apologies.
I am trying to recreate or re-implement OpenAI API , using the api and requests.
I am using
https://api.openai.com/v1/engines/code-davinci-002/completions
method: POST
I removed my token (for privacy purposes).
Optional<URLRequest>
▿ some : https://api.openai.com/v1/engines/code-davinci-002/completions
▿ url : Optional<URL>
▿ some : https://api.openai.com/v1/engines/code-davinci-002/completions
- _url : https://api.openai.com/v1/engines/code-davinci-002/completions
- cachePolicy : 0
- timeoutInterval : 60.0
- mainDocumentURL : nil
- networkServiceType : __C.NSURLRequestNetworkServiceType
- allowsCellularAccess : true
▿ httpMethod : Optional<String>
- some : "POST"
▿ allHTTPHeaderFields : Optional<Dictionary<String, String>>
▿ some : 2 elements
▿ 0 : 2 elements
- key : "Content-Type"
- value : "application/json"
▿ 1 : 2 elements
- key : "Authorization"
- value : "Bearer sk-<myToken_removed>"
▿ httpBody : Optional<Data>
▿ some : 414 bytes
- count : 414
▿ pointer : 0x00007ff464011400
- pointerValue : 140687626540032
- httpBodyStream : nil
- httpShouldHandleCookies : true
- httpShouldUsePipelining : false
static var sqlTranslate: Self {
.init(
urlString: "https://api.openai.com/v1/engines/code-davinci-002/completions",
method: .post,
httpBody: .completionDictionary(
prompt:
"""
### Postgres SQL tables, with their properties:
#
# Employee(id, name, department_id)
# Department(id, name, address)
# Salary_Payments(id, employee_id, amount, date)
#
### A query to list the names of the departments which employed more than 10 employees in the last 3 months
SELECT
"""
,
sampling: .temperature(0),
numberOfTokens: ...150,
stop: ["#", ";"],
user: nil,
presencePenalty: 0,
frequencyPenalty: 0,
bestOf: nil
)
)
}
It started working…
Perhaps there is a day long delay after approval for the api before it works?.. Anyway. it seems to be working.
Before working, I was getting the response discussed here: python - openai.error.InvalidRequestError: Engine not found - Stack Overflow
1 Like
sps
May 12, 2022, 6:55am
5
Yeah that seems to be the case. Glad it works now; I was having a hard time understanding the first block of code you shared.
Cheers!