Currently using "https://github.com/MacPaw/OpenAI/blob/main/README.md#chats"
Example Implementation:
let query = ChatQuery(model: .gpt3_5Turbo, messages: [
.init(role: .user, content: promptStatement),
], stream: true)
let result = try await openAI.chats(query: query)
return result.choices.first?.message.content ?? "ERROR OCCURRED"
}
Experience:
- I will occasionally get a response in as a completed response but typically receive an error response:
Error: dataCorrupted(Swift.DecodingError.Context(codingPath: [], debugDescription: "The given data was not valid JSON.", underlyingError: Optional(Error Domain=NSCocoaErrorDomain Code=3840 "Invalid value around line 1, column 0." UserInfo={NSDebugDescription=Invalid value around line 1, column 0., NSJSONSerializationErrorIndex=0})))
Question:
- Is stream supported for iOS app development. It appears it is from the documentation of this api but so far after many hours can't get it to do as i expect, the submission of a prompt will at most respond after 30seconds to a minute the complete response at best. I can never get a stream of tokens responded