add support for Custom API Endpoint

+llms/+internal/callOpenAIChatAPI.m

@@ -18,6 +18,7 @@
 %    - ResponseFormat (response_format)
 %    - Seed (seed)
 %    - ApiKey
+%    - Endpoint
 %    - TimeOut
 %    - StreamFun
 %   More details on the parameters: https://platform.openai.com/docs/api-reference/chat/create
@@ -48,7 +49,7 @@
 %   apiKey = "your-api-key-here"
 %
 %   % Send a request
-%   [text, message] = llms.internal.callOpenAIChatAPI(messages, functions, ApiKey=apiKey)
+%   [text, message] = llms.internal.callOpenAIChatAPI(messages, functions, ApiKey=apiKey, Endpoint=endpoint)
 
 %   Copyright 2023-2024 The MathWorks, Inc.
 
@@ -67,15 +68,16 @@
     nvp.ResponseFormat = "text"
     nvp.Seed = []
     nvp.ApiKey = ""
+    nvp.Endpoint = ""
     nvp.TimeOut = 10
     nvp.StreamFun = []
 end
 
-END_POINT = "https://api.openai.com/v1/chat/completions";
+// END_POINT = "https://api.openai.com/v1/chat/completions";
 
 parameters = buildParametersCall(messages, functions, nvp);
 
-[response, streamedText] = llms.internal.sendRequest(parameters,nvp.ApiKey, END_POINT, nvp.TimeOut, nvp.StreamFun);
+[response, streamedText] = llms.internal.sendRequest(parameters,nvp.ApiKey, nvp.Endpoint, nvp.TimeOut, nvp.StreamFun);
 
 % If call errors, "choices" will not be part of response.Body.Data, instead
 % we get response.Body.Data.error

+llms/+internal/getEndpointFromNvpOrEnv.m

@@ -0,0 +1,39 @@
+function endpoint = getEndpointFromNvpOrEnv(nvp, type='chat')
+% This function is undocumented and will change in a future release
+
+% getEndpointFromNvpOrEnv Retrieves an API key from a Name-Value Pair struct or environment variable.
+%
+%   This function takes a struct nvp containing name-value pairs and checks
+%   if it contains a field called "Endpoint". If the field is not found, 
+%   the function attempts to retrieve the API base_url from an environment
+%   variable called "OPENAI_API_BASE_URL". If both methods fail, the function 
+%   throws an error.
+
+%   Copyright 2023 The MathWorks, Inc.
+
+    openai_api_base_url = 'https://api.openai.com/v1'
+    if isfield(nvp, "Endpoint")
+        endpoint = nvp.Endpoint;
+    else
+        if isenv("OPENAI_API_BASE_URL")
+            base_url = getenv("OPENAI_API_BASE_URL");
+            if ~startsWith(base_url, 'http')
+                base_url = openai_api_base_url;
+            end
+        else
+            endpoint = openai_api_base_url;
+        end
+
+        completions.chat = "/chat/completions";
+        completions.embeddings = "/embeddings";
+        completions.image_generate = "/images/generations";
+        completions.image_edits = "/images/edits";
+        completions.image_variations = "/images/variations";
+        if ~isfield(completions, type)
+            endpoint = strcat(base_url, completions.chat);
+        else
+            endpoint = strcat(base_url, completions.(type));
+        end
+
+    end
+end

README.md

@@ -56,6 +56,7 @@ Set up your OpenAI API key. Create a `.env` file in the project root directory w
 
 ```
 OPENAI_API_KEY=<your key>
+OPENAI_API_BASE_URL=<your base_url> (optional, format like: "https://api.openai.com/v1")
 ```
 
 Then load your `.env` file as follows:

extractOpenAIEmbeddings.m

@@ -11,6 +11,8 @@
 %
 %   'ApiKey'                    - OpenAI API token. It can also be specified by
 %                                setting the environment variable OPENAI_API_KEY
+%   'Endpoint'                  - OpenAI API base_url. It can also be specified by
+%                                setting the environment variable OPENAI_API_BASE_URL
 %
 %   'TimeOut'                   - Connection Timeout in seconds (default: 10 secs)
 %
@@ -29,11 +31,13 @@
     nvp.TimeOut    (1,1) {mustBeReal,mustBePositive} = 10
     nvp.Dimensions (1,1) {mustBeInteger,mustBePositive}
     nvp.ApiKey           {llms.utils.mustBeNonzeroLengthTextScalar}
+    nvp.Endpoint   (1,1) string {mustBeTextScalar}
 end
 
-END_POINT = "https://api.openai.com/v1/embeddings";
+// END_POINT = "https://api.openai.com/v1/embeddings";
 
 key = llms.internal.getApiKeyFromNvpOrEnv(nvp);
+endpoint = llms.internal.getEndpointFromNvpOrEnv(nvp, 'embeddings');
 
 parameters = struct("input",text,"model",nvp.ModelName);
 
@@ -47,7 +51,7 @@
 end
 
 
-response = llms.internal.sendRequest(parameters,key, END_POINT, nvp.TimeOut);
+response = llms.internal.sendRequest(parameters,key, endpoint, nvp.TimeOut);
 
 if isfield(response.Body.Data, "data")
     emb = [response.Body.Data.data.embedding];

openAIChat.m

@@ -106,6 +106,7 @@
         Tools
         FunctionsStruct
         ApiKey
+        Endpoint
         StreamFun
     end
 
@@ -123,6 +124,7 @@
                 nvp.StopSequences                  {mustBeValidStop} = {}
                 nvp.ResponseFormat           (1,1) string {mustBeMember(nvp.ResponseFormat,["text","json"])} = "text"
                 nvp.ApiKey                         {mustBeNonzeroLengthTextScalar} 
+                nvp.Endpoint                 (1,1) string {mustBeTextScalar}
                 nvp.PresencePenalty                {mustBeValidPenalty} = 0
                 nvp.FrequencyPenalty               {mustBeValidPenalty} = 0
                 nvp.TimeOut                  (1,1) {mustBeReal,mustBePositive} = 10
@@ -171,6 +173,7 @@
             this.PresencePenalty = nvp.PresencePenalty;
             this.FrequencyPenalty = nvp.FrequencyPenalty;
             this.ApiKey = llms.internal.getApiKeyFromNvpOrEnv(nvp);
+            this.Endpoint = llms.internal.getEndpointFromNvpOrEnv(nvp, "chat");
             this.TimeOut = nvp.TimeOut;
         end
 
@@ -233,7 +236,7 @@
                 StopSequences=this.StopSequences, MaxNumTokens=nvp.MaxNumTokens, ...
                 PresencePenalty=this.PresencePenalty, FrequencyPenalty=this.FrequencyPenalty, ...
                 ResponseFormat=this.ResponseFormat,Seed=nvp.Seed, ...
-                ApiKey=this.ApiKey,TimeOut=this.TimeOut, StreamFun=this.StreamFun);
+                ApiKey=this.ApiKey,Endpoint=this.Endpoint,TimeOut=this.TimeOut, StreamFun=this.StreamFun);
 
             if isfield(response.Body.Data,"error")
                 err = response.Body.Data.error.message;

openAIImages.m

@@ -40,13 +40,15 @@
 
     properties (Access=private)
         ApiKey
+        Endpoint
     end
 
     methods
         function this = openAIImages(nvp)
             arguments
                 nvp.ModelName   (1,1) {mustBeMember(nvp.ModelName,["dall-e-2", "dall-e-3"])} = "dall-e-2"
                 nvp.ApiKey            {mustBeNonzeroLengthTextScalar} 
+                nvp.Endpoint    (1,1) string {mustBeTextScalar}
                 nvp.TimeOut     (1,1) {mustBeReal,mustBePositive} = 10
             end
 
@@ -96,7 +98,8 @@
                 nvp.Style               (1,1) string {mustBeMember(nvp.Style,["vivid", "natural"])} 
             end
 
-            endpoint = "https://api.openai.com/v1/images/generations";
+            // endpoint = "https://api.openai.com/v1/images/generations";
+            endpoint = llms.internal.getEndpointFromNvpOrEnv(nvp, "image_generate");
 
             validatePromptSize(this.ModelName, prompt)
             validateSizeNVP(this.ModelName, nvp.Size)
@@ -194,7 +197,8 @@
 
             validatePromptSize(this.ModelName, prompt)
 
-            endpoint = 'https://api.openai.com/v1/images/edits';
+            // endpoint = 'https://api.openai.com/v1/images/edits';
+            endpoint = llms.internal.getEndpointFromNvpOrEnv(nvp, "image_edits");
 
             % Required params
             numImages = num2str(nvp.NumImages);
@@ -252,7 +256,8 @@
                         this.ModelName));
             end
 
-            endpoint = 'https://api.openai.com/v1/images/variations';
+            // endpoint = 'https://api.openai.com/v1/images/variations';
+            endpoint = llms.internal.getEndpointFromNvpOrEnv(nvp, "image_variations");
 
             numImages = num2str(nvp.NumImages);
             body = matlab.net.http.io.MultipartFormProvider(...