[OpenAI] Add base url (#1757)

* [OpenAI] Add base url

* [OpenAI] Wip

bot/admin/server/src/test/kotlin/service/RAGServiceTest.kt

@@ -73,7 +73,8 @@ class RAGServiceTest : AbstractTest() {
                 apiKey = "apikey",
                 model = MODEL,
                 prompt = PROMPT,
-                temperature = TEMPERATURE
+                temperature = TEMPERATURE,
+                baseUrl = "https://api.openai.com/v1"
             ),
             emSetting = AzureOpenAIEMSettingDTO(
                 apiKey = "apiKey",

bot/admin/server/src/test/kotlin/service/RAGValidationServiceTest.kt

@@ -63,7 +63,8 @@ class RAGValidationServiceTest {
     }
 
     private val openAILLMSetting = OpenAILLMSetting(
-        apiKey = "123-abc", model = "unavailable-model", temperature = "0.4", prompt = "How to bike in the rain"
+        apiKey = "123-abc", model = "unavailable-model", temperature = "0.4", prompt = "How to bike in the rain",
+        baseUrl = "https://api.openai.com/v1",
     )
 
     private val azureOpenAIEMSetting = AzureOpenAIEMSettingDTO(

bot/admin/web/src/app/configuration/sentence-generation-settings/models/engines-configuration.ts

@@ -40,23 +40,24 @@ Answer in '{{locale}}' (language locale).
 
 export const EngineConfigurations: EnginesConfiguration[] = [
   {
-    label: 'OpenAi',
+    label: 'OpenAI',
     key: LLMProvider.OpenAI,
     params: [
       { key: 'apiKey', label: 'Api key', type: 'obfuscated' },
+      { key: 'baseUrl', label: 'Base url', type: 'text', defaultValue: 'https://api.openai.com/v1' },
       { key: 'model', label: 'Model name', type: 'openlist', source: OpenAIModelsList },
       { key: 'temperature', label: 'Temperature', type: 'number', inputScale: 'fullwidth' },
       { key: 'prompt', label: 'Prompt', type: 'prompt', inputScale: 'fullwidth', defaultValue: DefaultPrompt }
     ]
   },
   {
-    label: 'Azure OpenAi',
+    label: 'Azure OpenAI',
     key: LLMProvider.AzureOpenAIService,
     params: [
       { key: 'apiKey', label: 'Api key', type: 'obfuscated' },
       { key: 'apiVersion', label: 'Api version', type: 'openlist', source: AzureOpenAiApiVersionsList },
       { key: 'deploymentName', label: 'Deployment name', type: 'text' },
-      { key: 'apiBase', label: 'Private endpoint base url', type: 'obfuscated' },
+      { key: 'apiBase', label: 'Base url', type: 'obfuscated' },
       { key: 'temperature', label: 'Temperature', type: 'number', inputScale: 'fullwidth' },
       { key: 'prompt', label: 'Prompt', type: 'prompt', inputScale: 'fullwidth', defaultValue: DefaultPrompt }
     ]
@@ -65,7 +66,7 @@ export const EngineConfigurations: EnginesConfiguration[] = [
     label: 'Ollama',
     key: LLMProvider.Ollama,
     params: [
-      { key: 'baseUrl', label: 'BaseUrl', type: 'text', defaultValue: 'http://localhost:11434' },
+      { key: 'baseUrl', label: 'Base url', type: 'text', defaultValue: 'http://localhost:11434' },
       { key: 'model', label: 'Model', type: 'openlist', source: OllamaLlmModelsList, defaultValue: 'llama2' },
       { key: 'temperature', label: 'Temperature', type: 'number', inputScale: 'fullwidth', defaultValue: 0.7 },
       { key: 'prompt', label: 'Prompt', type: 'prompt', inputScale: 'fullwidth', defaultValue: DefaultPrompt }

bot/admin/web/src/app/rag/rag-settings/models/engines-configurations.ts

@@ -39,23 +39,24 @@ Answer in {locale}.
 
 const EnginesConfigurations_Llm: EnginesConfiguration[] = [
   {
-    label: 'OpenAi',
+    label: 'OpenAI',
     key: LLMProvider.OpenAI,
     params: [
       { key: 'apiKey', label: 'Api key', type: 'obfuscated' },
+      { key: 'baseUrl', label: 'Base url', type: 'text', defaultValue: 'https://api.openai.com/v1' },
       { key: 'model', label: 'Model name', type: 'openlist', source: OpenAIModelsList },
       { key: 'temperature', label: 'Temperature', type: 'number', inputScale: 'fullwidth' },
       { key: 'prompt', label: 'Prompt', type: 'prompt', inputScale: 'fullwidth', defaultValue: DefaultPrompt }
     ]
   },
   {
-    label: 'Azure OpenAi',
+    label: 'Azure OpenAI',
     key: LLMProvider.AzureOpenAIService,
     params: [
       { key: 'apiKey', label: 'Api key', type: 'obfuscated' },
       { key: 'apiVersion', label: 'Api version', type: 'openlist', source: AzureOpenAiApiVersionsList },
       { key: 'deploymentName', label: 'Deployment name', type: 'text' },
-      { key: 'apiBase', label: 'Private endpoint base url', type: 'obfuscated' },
+      { key: 'apiBase', label: 'Base url', type: 'obfuscated' },
       { key: 'temperature', label: 'Temperature', type: 'number', inputScale: 'fullwidth' },
       { key: 'prompt', label: 'Prompt', type: 'prompt', inputScale: 'fullwidth', defaultValue: DefaultPrompt }
     ]
@@ -74,21 +75,22 @@ const EnginesConfigurations_Llm: EnginesConfiguration[] = [
 
 const EnginesConfigurations_Embedding: EnginesConfiguration[] = [
   {
-    label: 'OpenAi',
+    label: 'OpenAI',
     key: LLMProvider.OpenAI,
     params: [
       { key: 'apiKey', label: 'Api key', type: 'obfuscated' },
+      { key: 'baseUrl', label: 'Base url', type: 'text', defaultValue: 'https://api.openai.com/v1' },
       { key: 'model', label: 'Model name', type: 'openlist', source: OpenAIEmbeddingModel }
     ]
   },
   {
-    label: 'Azure OpenAi',
+    label: 'Azure OpenAI',
     key: LLMProvider.AzureOpenAIService,
     params: [
       { key: 'apiKey', label: 'Api key', type: 'obfuscated' },
       { key: 'apiVersion', label: 'Api version', type: 'openlist', source: AzureOpenAiApiVersionsList },
       { key: 'deploymentName', label: 'Deployment name', type: 'text' },
-      { key: 'apiBase', label: 'Private endpoint base url', type: 'obfuscated' }
+      { key: 'apiBase', label: 'Base url', type: 'obfuscated' }
     ]
   },
   {

bot/admin/web/src/app/shared/model/ai-settings.ts

@@ -54,12 +54,16 @@ export const AzureOpenAiApiVersionsList: string[] = [
 ];
 
 export const OpenAIModelsList: string[] = [
+  'gpt-4o',
+  'gpt-4o-mini',
+
   'gpt-4',
   'gpt-4-0314',
   'gpt-4-0613',
   'gpt-4-32k',
   'gpt-4-32k-0314',
   'gpt-4-32k-0613',
+  'gpt-4-turbo',
 
   'gpt-3.5-turbo',
   'gpt-3.5-turbo-0613',
@@ -71,7 +75,7 @@ export const OpenAIModelsList: string[] = [
   'davinci-002'
 ];
 
-export const OpenAIEmbeddingModel: string[] = ['text-embedding-ada-002'];
+export const OpenAIEmbeddingModel: string[] = ['text-embedding-3-small', 'text-embedding-3-large', 'text-embedding-ada-002'];
 
 export const OllamaLlmModelsList: string[] = ['llama2', 'llama3', 'llama3.1', 'llama3.1:8b', 'llama3.2'];
 

bot/storage-mongo/src/test/kotlin/BotRAGConfigurationMongoDAOTest.kt

@@ -48,11 +48,13 @@ internal class BotRAGConfigurationMongoDAOTest : AbstractTest() {
                 apiKey = RawSecretKey("apiKey1"),
                 model = "modelName1",
                 temperature = "1F",
-                prompt = "prompt1"
+                prompt = "prompt1",
+                baseUrl = "https://api.openai.com/v1"
             ),
             emSetting = OpenAIEMSetting(
                 apiKey = RawSecretKey("apiKey1"),
-                model = "modelName1"
+                model = "modelName1",
+                baseUrl = "https://api.openai.com/v1"
             ),
             noAnswerSentence = "no answer sentence"
         )
@@ -73,11 +75,13 @@ internal class BotRAGConfigurationMongoDAOTest : AbstractTest() {
             llmSetting = OpenAILLMSetting(
                 apiKey = RawSecretKey("apiKey1"),
                 model = "modelName1",
+                baseUrl = "https://api.openai.com/v1",
                 temperature = "1F",
                 prompt = "prompt1"
             ),
             emSetting = OpenAIEMSetting(
                 apiKey = RawSecretKey("apiKey1"),
+                baseUrl = "https://api.openai.com/v1",
                 model = "modelName1"
             ),
             noAnswerSentence = "no answer sentence1"
@@ -92,10 +96,12 @@ internal class BotRAGConfigurationMongoDAOTest : AbstractTest() {
                 apiKey = RawSecretKey("apiKey1"),
                 model = "modelName1",
                 temperature = "1F",
+                baseUrl = "https://api.openai.com/v1",
                 prompt = "prompt1"
             ),
             emSetting = OpenAIEMSetting(
                 apiKey = RawSecretKey("apiKey1"),
+                baseUrl = "https://api.openai.com/v1",
                 model = "modelName1"
             ),
             noAnswerSentence = "no answer sentence1"
@@ -124,10 +130,12 @@ internal class BotRAGConfigurationMongoDAOTest : AbstractTest() {
                 apiKey = RawSecretKey("apiKey1"),
                 model = "modelName1",
                 temperature = "1F",
+                baseUrl = "https://api.openai.com/v1",
                 prompt = "prompt1"
             ),
             emSetting = OpenAIEMSetting(
                 apiKey = RawSecretKey("apiKey1"),
+                baseUrl = "https://api.openai.com/v1",
                 model = "modelName1"
             ),
             noAnswerSentence = "no answer sentence"

gen-ai/orchestrator-core/src/main/kotlin/ai/tock/genai/orchestratorcore/mappers/EMSettingMapper.kt

@@ -34,11 +34,20 @@ object EMSettingMapper {
         with(entity){
             when(this){
                 is OpenAIEMSetting ->
-                    OpenAIEMSetting(SecurityUtils.fetchSecretKeyValue(apiKey), model)
+                    OpenAIEMSetting(
+                        apiKey = SecurityUtils.fetchSecretKeyValue(apiKey),
+                        model = model,
+                        baseUrl = baseUrl
+                    )
                 is AzureOpenAIEMSetting ->
-                    AzureOpenAIEMSetting(SecurityUtils.fetchSecretKeyValue(apiKey), apiBase, deploymentName, apiVersion)
+                    AzureOpenAIEMSetting(
+                        apiKey = SecurityUtils.fetchSecretKeyValue(apiKey),
+                        apiBase = apiBase,
+                        deploymentName = deploymentName,
+                        apiVersion = apiVersion
+                    )
                 is OllamaEMSetting ->
-                    OllamaEMSetting(model, baseUrl)
+                    OllamaEMSetting(model = model, baseUrl = baseUrl)
                 else ->
                     throw IllegalArgumentException("Unsupported EM Setting")
             }
@@ -56,16 +65,20 @@ object EMSettingMapper {
         with(dto){
             when(this){
                 is OpenAIEMSetting ->
-                    OpenAIEMSetting(SecurityUtils.createSecretKey(namespace, botId, feature, apiKey), model)
+                    OpenAIEMSetting(
+                        apiKey = SecurityUtils.createSecretKey(namespace, botId, feature, apiKey),
+                        model = model,
+                        baseUrl = baseUrl
+                    )
                 is AzureOpenAIEMSetting ->
                     AzureOpenAIEMSetting(
                         SecurityUtils.createSecretKey(namespace, botId, feature, apiKey),
-                        apiBase,
-                        deploymentName,
-                        apiVersion
+                        apiBase = apiBase,
+                        deploymentName = deploymentName,
+                        apiVersion = apiVersion
                     )
                 is OllamaEMSetting ->
-                    OllamaEMSetting(model, baseUrl)
+                    OllamaEMSetting(model = model, baseUrl = baseUrl)
                 else ->
                     throw IllegalArgumentException("Unsupported EM Setting")
             }

gen-ai/orchestrator-core/src/main/kotlin/ai/tock/genai/orchestratorcore/mappers/LLMSettingMapper.kt

@@ -34,18 +34,29 @@ object LLMSettingMapper {
 
             when (this) {
                 is OpenAILLMSetting ->
-                    OpenAILLMSetting(SecurityUtils.fetchSecretKeyValue(apiKey), temperature, prompt, model)
+                    OpenAILLMSetting(
+                        apiKey = SecurityUtils.fetchSecretKeyValue(apiKey),
+                        temperature = temperature,
+                        prompt = prompt,
+                        model = model,
+                        baseUrl = baseUrl
+                    )
                 is AzureOpenAILLMSetting ->
                     AzureOpenAILLMSetting(
-                        SecurityUtils.fetchSecretKeyValue(apiKey),
-                        temperature,
-                        prompt,
-                        apiBase,
-                        deploymentName,
-                        apiVersion
+                        apiKey = SecurityUtils.fetchSecretKeyValue(apiKey),
+                        temperature = temperature,
+                        prompt = prompt,
+                        apiBase = apiBase,
+                        deploymentName = deploymentName,
+                        apiVersion = apiVersion
                     )
                 is OllamaLLMSetting ->
-                    OllamaLLMSetting(temperature, prompt, model, baseUrl)
+                    OllamaLLMSetting(
+                        temperature = temperature,
+                        prompt = prompt,
+                        model = model,
+                        baseUrl = baseUrl
+                    )
                 else ->
                     throw IllegalArgumentException("Unsupported LLM Setting")
             }
@@ -64,10 +75,11 @@ object LLMSettingMapper {
             when (this) {
                 is OpenAILLMSetting ->
                     OpenAILLMSetting(
-                        SecurityUtils.createSecretKey(namespace, botId, feature, apiKey),
-                        temperature,
-                        prompt,
-                        model
+                        apiKey = SecurityUtils.createSecretKey(namespace, botId, feature, apiKey),
+                        temperature = temperature,
+                        prompt = prompt,
+                        model = model,
+                        baseUrl = baseUrl
                     )
                 is AzureOpenAILLMSetting ->
                     AzureOpenAILLMSetting(

gen-ai/orchestrator-core/src/main/kotlin/ai/tock/genai/orchestratorcore/models/em/OpenAIEMSetting.kt

@@ -19,6 +19,7 @@ package ai.tock.genai.orchestratorcore.models.em
 data class OpenAIEMSetting<T>(
     override val apiKey: T,
     val model: String,
+    val baseUrl: String,
 ) : EMSettingBase<T>(EMProvider.OpenAI, apiKey)
 
 typealias OpenAIEMSettingDTO = OpenAIEMSetting<String>

gen-ai/orchestrator-core/src/main/kotlin/ai/tock/genai/orchestratorcore/models/llm/OpenAILLMSetting.kt

@@ -21,6 +21,7 @@ data class OpenAILLMSetting<T>(
     override val temperature: String,
     override val prompt: String,
     val model: String,
+    val baseUrl: String,
 ) : LLMSettingBase<T>(LLMProvider.OpenAI, apiKey, temperature, prompt) {
     override fun copyWithTemperature(temperature: String): LLMSettingBase<T> {
         return this.copy(temperature=temperature)

gen-ai/orchestrator-server/src/main/python/server/src/gen_ai_orchestrator/models/em/openai/openai_em_setting.py

@@ -38,3 +38,8 @@ class OpenAIEMSetting(BaseEMSetting):
         examples=[RawSecretKey(value='ab7-14Ed2-dfg2F-A1IV4B')]
     )
     model: str = Field(description='The model id', examples=['text-embedding-ada-002'])
+    base_url: str = Field(
+        description='The OpenAI endpoint base URL',
+        examples=["https://api.openai.com/v1"],
+        default="https://api.openai.com/v1"
+    )

gen-ai/orchestrator-server/src/main/python/server/src/gen_ai_orchestrator/models/llm/openai/openai_llm_setting.py

@@ -40,3 +40,8 @@ class OpenAILLMSetting(BaseLLMSetting):
     model: str = Field(
         description='The model id', examples=['gpt-3.5-turbo'], min_length=1
     )
+    base_url: str = Field(
+        description='The OpenAI endpoint base URL',
+        examples=["https://api.openai.com/v1"],
+        default="https://api.openai.com/v1"
+    )

gen-ai/orchestrator-server/src/main/python/server/src/gen_ai_orchestrator/services/langchain/factories/em/openai_em_factory.py

@@ -17,6 +17,7 @@
 
 from langchain.embeddings.base import Embeddings
 from langchain_openai import OpenAIEmbeddings
+from openai import base_url
 
 from gen_ai_orchestrator.configurations.environment.settings import (
     application_settings,
@@ -46,6 +47,7 @@ class OpenAIEMFactory(LangChainEMFactory):
     def get_embedding_model(self) -> Embeddings:
         return OpenAIEmbeddings(
             openai_api_key=fetch_secret_key_value(self.setting.api_key),
+            base_url=self.setting.base_url,
             model=self.setting.model,
             timeout=application_settings.em_provider_timeout,
         )

gen-ai/orchestrator-server/src/main/python/server/src/gen_ai_orchestrator/services/langchain/factories/llm/openai_llm_factory.py

@@ -48,6 +48,7 @@ class OpenAILLMFactory(LangChainLLMFactory):
     def get_language_model(self) -> BaseLanguageModel:
         return ChatOpenAI(
             openai_api_key=fetch_secret_key_value(self.setting.api_key),
+            base_url=self.setting.base_url,
             model_name=self.setting.model,
             temperature=self.setting.temperature,
             request_timeout=application_settings.llm_provider_timeout,