How to use Webwrite to call REST APIs successfully, eg OpenAI API

Webwrite Produces Error Though Seemingly Identical CURL Request Is Successful at Calling OpenAI API

This matlab code doesn't work. Can anyone figure out why?

prompt = 'Hello, how are you today?';

data = struct('prompt', prompt, 'temperature', 0.7, 'max_tokens', 60);

options = weboptions('RequestMethod','POST','MediaType','application/json','KeyName','Authorization','KeyValue','Bearer xxx');

response = webwrite(url,data,options)

The code generates a server error:

The server returned the status 400 with message "Bad Request" in response to the request to URL

https://api.openai.com/v1/completions.

This code was based on a working CURL example below that successfully calls the openAI API.

In each case, the Authorization xxx is replaced with a functional openAI api key.

Working CURL:

## cURL

curl -X "POST" "https://api.openai.com/v1/chat/completions" \

-H 'Content-Type: application/json' \

-H 'Authorization: Bearer xxxxx' \

-d $'{

"model": "gpt-3.5-turbo",

"messages": [

{

"content": "Hello!",

"role": "user"

}

]

}'

This is running on a Mac with up to date OS. This is true using Matlab 2022b or 2023a prerelease.

Thank you for your help

Answer 1

[CdC]

Thank you for the answer above. Based on it, I created the following function. This is both a working example of how to call rest apis, and a working example of how to call the openAI Chat api. It remains unclear to me why we need to use matlab.net and why webwrite doesn't work, given that this would apepar to be webrite's function.

function [result,tokens] = openAIChat(content)
%openAIChat Call openAI REST API for chat completions
%   Detailed explanation goes here
tic
import matlab.net.*
import matlab.net.http.*

% Define the API endpoint Davinci
api_endpoint = "https://api.openai.com/v1/chat/completions";

% Define the API key from https://beta.openai.com/account/api-keys
api_key = "xxxxxxx";

data = jsondecode('{"model":"gpt-3.5-turbo","messages":[{"role":"system","content":"You are a helpful assistant."},{"role":"user","content":"What is my birthday?"}]}');
data.messages(2).content = content;
% Define the headers for the API request
headers = matlab.net.http.HeaderField('Content-Type', 'application/json');
headers(2) = matlab.net.http.HeaderField('Authorization', 'Bearer ' + api_key);
% Define the request message
request = matlab.net.http.RequestMessage('post',headers,data);

% Send the request and store the response
response = send(request, URI(api_endpoint));

% Extract the response text
result = response.Body.Data.choices.message.content;
tokens = response.Body.Data.usage;
%disp(result);
end

[Radoslav Kriz]

Its good job. THX

Answer 2

[Toshiaki Takeuchi]

Here is a MATLAB app that provides sample prompts as presets, and you can run chat-generated MATLAB code inside the app. People often struggle with prompt engineering, and the presets gives you a good starting point.

https://www.mathworks.com/matlabcentral/fileexchange/126665-matgpt

You can either download it from GitHub into your MATLAB path or use "Open in MATLAB Online" to import the repo into MATLAB Online. Once you have the API key, you can set it up in the OS environment variable.

setenv("OPENAI_API_KEY","your key here")

Finally, double click on MatGPT.mlapp to open it in App Designer and run it there.

If you would rather use command line interface, you can use chatGPT class included in the repo.

Answer 3

[the cyclist]

I did not carefully look at your code, but there was recently discussion at this question, which reached a successful conclusion (including pointers to a couple File Exchange submission).

I expect it will be helpful toward goal.

[CdC]

Thanks for your help. Please note posted code based on your answer and link.