Add new model type to use custom OpenAI api compatible servers

gpt4all-chat/chatgpt.cpp

@@ -154,7 +154,7 @@ void ChatGPTWorker::request(const QString &apiKey,
 {
     m_ctx = promptCtx;
 
-    QUrl openaiUrl("https://api.openai.com/v1/chat/completions");
+    QUrl openaiUrl(m_chat->APIBase() + "/chat/completions");
     const QString authorization = QString("Bearer %1").arg(apiKey).trimmed();
     QNetworkRequest request(openaiUrl);
     request.setHeader(QNetworkRequest::ContentTypeHeader, "application/json");

gpt4all-chat/chatgpt.h

@@ -63,8 +63,11 @@ class ChatGPT : public QObject, public LLModel {
     void setThreadCount(int32_t n_threads) override;
     int32_t threadCount() const override;
 
+    const QString& APIBase() const { return m_apiBase; }
+
     void setModelName(const QString &modelName) { m_modelName = modelName; }
     void setAPIKey(const QString &apiKey) { m_apiKey = apiKey; }
+    void setAPIBase(const QString &apiBase) { m_apiBase = apiBase; }
 
     QList<QString> context() const { return m_context; }
     void setContext(const QList<QString> &context) { m_context = context; }
@@ -91,6 +94,7 @@ class ChatGPT : public QObject, public LLModel {
     std::function<bool(int32_t, const std::string&)> m_responseCallback;
     QString m_modelName;
     QString m_apiKey;
+    QString m_apiBase;
     QList<QString> m_context;
 };
 

gpt4all-chat/chatllm.cpp

@@ -156,7 +156,7 @@ bool ChatLLM::loadModel(const ModelInfo &modelInfo)
     if (isModelLoaded() && this->modelInfo() == modelInfo)
         return true;
 
-    bool isChatGPT = modelInfo.isChatGPT;
+    bool isChatGPT = modelInfo.isChatGPT || modelInfo.isOpenAICompatible;
     QString filePath = modelInfo.dirpath + modelInfo.filename();
     QFileInfo fileInfo(filePath);
 
@@ -233,18 +233,32 @@ bool ChatLLM::loadModel(const ModelInfo &modelInfo)
     if (fileInfo.exists()) {
         if (isChatGPT) {
             QString apiKey;
-            QString chatGPTModel = fileInfo.completeBaseName().remove(0, 8); // remove the chatgpt- prefix
+            QString apiBase;
+            QString chatGPTModel;
             {
                 QFile file(filePath);
                 file.open(QIODeviceBase::ReadOnly | QIODeviceBase::Text);
                 QTextStream stream(&file);
-                apiKey = stream.readAll();
+                QStringList chatGPTParams = stream.readAll().split('\n');
+                if (chatGPTParams.isEmpty()) {
+
+                    emit modelLoadingError(QString("Could not load model due to invalid settings for %1").arg(modelInfo.filename()));
+                } else {
+                    apiKey = chatGPTParams[0];
+                    if (chatGPTParams.size() >= 2) {
+                        apiBase = chatGPTParams[1];
+                    }
+                    if (chatGPTParams.size() >= 3) {
+i                       chatGPTModel = chatGPTParams[2];
+                    }
+                }
                 file.close();
             }
             m_llModelType = LLModelType::CHATGPT_;
             ChatGPT *model = new ChatGPT();
-            model->setModelName(chatGPTModel);
+            model->setModelName(chatGPTModel.size() ? chatGPTModel : fileInfo.completeBaseName().remove(0, 8)); // remove the chatgpt- prefix
             model->setAPIKey(apiKey);
+            model->setAPIBase(apiBase.size() ? apiBase : "https://api.openai.com/v1/");
             m_llModelInfo.model = model;
         } else {
 

gpt4all-chat/download.cpp

@@ -172,18 +172,21 @@ void Download::cancelDownload(const QString &modelFile)
     }
 }
 
-void Download::installModel(const QString &modelFile, const QString &apiKey)
+void Download::installModel(const QString &modelFile, const QString &apiBase, const QString &apiKey, const QString &modelName)
 {
-    Q_ASSERT(!apiKey.isEmpty());
-    if (apiKey.isEmpty())
+    Q_ASSERT(!(apiKey.isEmpty() && apiBase.isEmpty()));
+    if (apiKey.isEmpty() && apiBase.isEmpty())
         return;
 
     Network::globalInstance()->sendInstallModel(modelFile);
     QString filePath = MySettings::globalInstance()->modelPath() + modelFile;
     QFile file(filePath);
     if (file.open(QIODeviceBase::WriteOnly | QIODeviceBase::Text)) {
         QTextStream stream(&file);
-        stream << apiKey;
+        stream << apiKey + "\n" + apiBase;
+        if (modelName.size()) {
+            stream << "\n" + modelName;
+        }
         file.close();
     }
 }

gpt4all-chat/download.h

@@ -52,7 +52,7 @@ class Download : public QObject
     bool hasNewerRelease() const;
     Q_INVOKABLE void downloadModel(const QString &modelFile);
     Q_INVOKABLE void cancelDownload(const QString &modelFile);
-    Q_INVOKABLE void installModel(const QString &modelFile, const QString &apiKey);
+    Q_INVOKABLE void installModel(const QString &modelFile, const QString &apiBase, const QString &apiKey, const QString &modelName = "");
     Q_INVOKABLE void removeModel(const QString &modelFile);
     Q_INVOKABLE bool isFirstStart() const;
 

gpt4all-chat/main.qml

@@ -967,7 +967,7 @@ Window {
                 }
 
                 Image {
-                    visible: currentChat.isServer || currentChat.modelInfo.isChatGPT
+                    visible: currentChat.isServer || currentChat.modelInfo.isChatGPT || currentChat.modelInfo.isOpenAICompatible
                     anchors.fill: parent
                     sourceSize.width: 1024
                     sourceSize.height: 1024

gpt4all-chat/modellist.cpp

@@ -356,7 +356,7 @@ ModelInfo ModelList::defaultModelInfo() const
         const size_t ramrequired = defaultModel->ramrequired;
 
         // If we don't have either setting, then just use the first model that requires less than 16GB that is installed
-        if (!hasUserDefaultName && !info->isChatGPT && ramrequired > 0 && ramrequired < 16)
+        if (!hasUserDefaultName && !(info->isChatGPT || info->isOpenAICompatible) && ramrequired > 0 && ramrequired < 16)
             break;
 
         // If we have a user specified default and match, then use it
@@ -479,6 +479,8 @@ QVariant ModelList::dataInternal(const ModelInfo *info, int role) const
             return info->isDefault;
         case ChatGPTRole:
             return info->isChatGPT;
+        case OpenAICompatibleRole:
+            return info->isOpenAICompatible;
         case DisableGUIRole:
             return info->disableGUI;
         case DescriptionRole:
@@ -604,6 +606,8 @@ void ModelList::updateData(const QString &id, int role, const QVariant &value)
             info->isDefault = value.toBool(); break;
         case ChatGPTRole:
             info->isChatGPT = value.toBool(); break;
+        case OpenAICompatibleRole:
+            info->isOpenAICompatible = value.toBool(); break;
         case DisableGUIRole:
             info->disableGUI = value.toBool(); break;
         case DescriptionRole:
@@ -735,6 +739,7 @@ QString ModelList::clone(const ModelInfo &model)
     updateData(id, ModelList::DirpathRole, model.dirpath);
     updateData(id, ModelList::InstalledRole, model.installed);
     updateData(id, ModelList::ChatGPTRole, model.isChatGPT);
+    updateData(id, ModelList::OpenAICompatibleRole, model.isOpenAICompatible);
     updateData(id, ModelList::TemperatureRole, model.temperature());
     updateData(id, ModelList::TopPRole, model.topP());
     updateData(id, ModelList::TopKRole, model.topK());
@@ -875,7 +880,9 @@ void ModelList::updateModelsFromDirectory()
 
                     for (const QString &id : modelsById) {
                         updateData(id, FilenameRole, filename);
-                        updateData(id, ChatGPTRole, filename.startsWith("chatgpt-"));
+                        const bool is_chatgpt_custom = filename.startsWith("chatgpt-custom");
+                        updateData(id, OpenAICompatibleRole, is_chatgpt_custom);
+                        updateData(id, ChatGPTRole, (!is_chatgpt_custom && filename.startsWith("chatgpt-")));
                         updateData(id, DirpathRole, info.dir().absolutePath() + "/");
                         updateData(id, FilesizeRole, toFileSize(info.size()));
                     }
@@ -1116,6 +1123,35 @@ void ModelList::parseModelsJsonFile(const QByteArray &jsonData, bool save)
             updateData(id, ModelList::SystemPromptRole, obj["systemPrompt"].toString());
     }
 
+    const QString CustomOpenAIDesc = tr("<ul><li>Requires acces to OpenAI compatible server</li>"
+                                        "<li>Any OpenAI compatible server that support chat/completions method</li></li></li>"
+                                        "<br />"
+                                        "<p>For OpenAI, the base path is \"https://api.openai.com/v1/\"</p>"
+                                        "<p>eg: For a selfhosted server it could be \"http://localhost:8008/v1/\"</p>");
+
+    {
+        const QString modelName = "Custom OpenAI API";
+        const QString id = modelName;
+        const QString modelFilename = "chatgpt-custom.txt";
+        if (contains(modelFilename))
+            changeId(modelFilename, id);
+        if (!contains(id))
+            addModel(id);
+        updateData(id, ModelList::NameRole, modelName);
+        updateData(id, ModelList::FilenameRole, modelFilename);
+        updateData(id, ModelList::FilesizeRole, "minimal");
+        updateData(id, ModelList::ChatGPTRole, false);
+        updateData(id, ModelList::OpenAICompatibleRole, true);
+        updateData(id, ModelList::DescriptionRole,
+            tr("<strong>Any OpenAI compatible server</strong><br>") + CustomOpenAIDesc);
+        updateData(id, ModelList::RequiresVersionRole, "2.4.2");
+        updateData(id, ModelList::OrderRole, "ca");
+        updateData(id, ModelList::RamrequiredRole, 0);
+        updateData(id, ModelList::ParametersRole, "?");
+        updateData(id, ModelList::QuantRole, "NA");
+        updateData(id, ModelList::TypeRole, "GPT");
+    }
+
     const QString chatGPTDesc = tr("<ul><li>Requires personal OpenAI API key.</li><li>WARNING: Will send"
         " your chats to OpenAI!</li><li>Your API key will be stored on disk</li><li>Will only be used"
         " to communicate with OpenAI</li><li>You can apply for an API key"
@@ -1133,6 +1169,7 @@ void ModelList::parseModelsJsonFile(const QByteArray &jsonData, bool save)
         updateData(id, ModelList::FilenameRole, modelFilename);
         updateData(id, ModelList::FilesizeRole, "minimal");
         updateData(id, ModelList::ChatGPTRole, true);
+        updateData(id, ModelList::OpenAICompatibleRole, false);
         updateData(id, ModelList::DescriptionRole,
             tr("<strong>OpenAI's ChatGPT model GPT-3.5 Turbo</strong><br>") + chatGPTDesc);
         updateData(id, ModelList::RequiresVersionRole, "2.4.2");
@@ -1157,6 +1194,7 @@ void ModelList::parseModelsJsonFile(const QByteArray &jsonData, bool save)
         updateData(id, ModelList::FilenameRole, modelFilename);
         updateData(id, ModelList::FilesizeRole, "minimal");
         updateData(id, ModelList::ChatGPTRole, true);
+        updateData(id, ModelList::OpenAICompatibleRole, false);
         updateData(id, ModelList::DescriptionRole,
             tr("<strong>OpenAI's ChatGPT model GPT-4</strong><br>") + chatGPTDesc + chatGPT4Warn);
         updateData(id, ModelList::RequiresVersionRole, "2.4.2");

gpt4all-chat/modellist.h

@@ -17,6 +17,7 @@ struct ModelInfo {
     Q_PROPERTY(bool isDefault MEMBER isDefault)
     Q_PROPERTY(bool disableGUI MEMBER disableGUI)
     Q_PROPERTY(bool isChatGPT MEMBER isChatGPT)
+    Q_PROPERTY(bool isOpenAICompatible MEMBER isOpenAICompatible)
     Q_PROPERTY(QString description MEMBER description)
     Q_PROPERTY(QString requiresVersion MEMBER requiresVersion)
     Q_PROPERTY(QString deprecatedVersion MEMBER deprecatedVersion)
@@ -61,6 +62,7 @@ struct ModelInfo {
     bool installed = false;
     bool isDefault = false;
     bool isChatGPT = false;
+    bool isOpenAICompatible = false;
     bool disableGUI = false;
     QString description;
     QString requiresVersion;
@@ -204,6 +206,7 @@ class ModelList : public QAbstractListModel
         InstalledRole,
         DefaultRole,
         ChatGPTRole,
+        OpenAICompatibleRole,
         DisableGUIRole,
         DescriptionRole,
         RequiresVersionRole,
@@ -246,6 +249,7 @@ class ModelList : public QAbstractListModel
         roles[InstalledRole] = "installed";
         roles[DefaultRole] = "isDefault";
         roles[ChatGPTRole] = "isChatGPT";
+        roles[OpenAICompatibleRole] = "isOpenAICompatible";
         roles[DisableGUIRole] = "disableGUI";
         roles[DescriptionRole] = "description";
         roles[RequiresVersionRole] = "requiresVersion";