diff --git a/client/src/copilot/generators/suggestionProvider.ts b/client/src/copilot/generators/suggestionProvider.ts
index 2d5a699..04e787f 100644
--- a/client/src/copilot/generators/suggestionProvider.ts
+++ b/client/src/copilot/generators/suggestionProvider.ts
@@ -2,7 +2,7 @@ import * as vscode from 'vscode';
 import { LanguageClient } from 'vscode-languageclient/browser';
 import { log } from '../../log';
 import { Documents, ModelConfig, PromptConfig } from '../utils/types';
-import { DEFAULT_LLM_MODELS, DEFAULT_LLM_ENDPOINTS } from '../utils/constants';
+import { DEFAULT_LLM_ENDPOINTS } from '../utils/constants';
 import { setLLMHealthStatus } from '../healthCheck';
 
 export async function getSuggestion(client: LanguageClient, documents: Documents, promptConfig: PromptConfig): Promise<string | null> {
@@ -10,26 +10,20 @@ export async function getSuggestion(client: LanguageClient, documents: Documents
     const apiKey = config.get<string>('apiKey');
     const provider = config.get<string>('provider');
     let llmModel = config.get<string>('llmModel');
-    let apiUrl;
 
-    if (!llmModel) {
-        switch (provider) {
-            case 'gemini':
-                llmModel = DEFAULT_LLM_MODELS.GEMINI;
-                apiUrl =  DEFAULT_LLM_ENDPOINTS.GEMINI;
-                break;
-            case 'openai':
-                llmModel = DEFAULT_LLM_MODELS.OPENAI;
-                apiUrl =  DEFAULT_LLM_ENDPOINTS.OPENAI;
-                break;
-            case 'mistralai':
-                llmModel = DEFAULT_LLM_MODELS.MISTRALAI;
-                apiUrl =  DEFAULT_LLM_ENDPOINTS.MISTRALAI;
-                break;   
-            default:
-                llmModel = '';
-                apiUrl = '';
-        }
+    let apiUrl;
+    switch (provider) {
+        case 'gemini':
+            apiUrl =  DEFAULT_LLM_ENDPOINTS.GEMINI;
+            break;
+        case 'openai':
+            apiUrl =  DEFAULT_LLM_ENDPOINTS.OPENAI;
+            break;
+        case 'mistral':
+            apiUrl =  DEFAULT_LLM_ENDPOINTS.MISTRALAI;
+            break;
+        default:
+            apiUrl = '';
     }
 
     // keys like maxTokens, temperature, topP comes from additionalParams
@@ -40,16 +34,9 @@ export async function getSuggestion(client: LanguageClient, documents: Documents
         llmModel,
         apiUrl,
         accessToken: apiKey,
-        additionalParams: {} 
+        additionalParams: {...additionalParams}
     };
 
-    if (additionalParams) {
-        modelConfig.additionalParams = {
-            ...modelConfig.additionalParams,
-            ...additionalParams
-        };
-    }
-
     try {
         log('Generating content...');
         const response: string = await client.sendRequest('generateContent', {
diff --git a/client/src/copilot/modelGeneratorWizard/modelGeneratorPanel.ts b/client/src/copilot/modelGeneratorWizard/modelGeneratorPanel.ts
index f0c2a4a..db81f99 100644
--- a/client/src/copilot/modelGeneratorWizard/modelGeneratorPanel.ts
+++ b/client/src/copilot/modelGeneratorWizard/modelGeneratorPanel.ts
@@ -40,8 +40,8 @@ export function createFileGeneratorPanel(context: vscode.ExtensionContext, clien
     });
 
     currentPanel.onDidChangeViewState(e => {
-        if (currentPanel) {
-            preloadFileLists(currentPanel, context);
+        if (currentPanel.visible) {
+            preloadFileLists(currentPanel);
         }
     });
 
@@ -53,9 +53,11 @@ export function createFileGeneratorPanel(context: vscode.ExtensionContext, clien
                     try {
                         await generateGrammarFile(client, message.filePath);
                         updateGeneratingState('grammar', false);
+                        vscode.window.showInformationMessage('Grammar file generated successfully');
                     } catch (error) {
                         log(`Error generating grammar file: ${error.message}`);
                         updateGeneratingState('grammar', false, error.message);
+                        vscode.window.showErrorMessage('Error generating grammar file');
                     }
                     break;
                 case FILE_GENERATORS.GENERATE_MODEL_FILE:
@@ -63,9 +65,11 @@ export function createFileGeneratorPanel(context: vscode.ExtensionContext, clien
                     try {
                         await generateModelFile(client, message.packageFile, message.grammarFile);
                         updateGeneratingState('model', false);
+                        vscode.window.showInformationMessage('Model file generated successfully');
                     } catch (error) {
                         log(`Error generating model file: ${error.message}`);
                         updateGeneratingState('model', false, error.message);
+                        vscode.window.showErrorMessage('Error generating model file');
                     }
                     break;
                 case FILE_GENERATORS.REQUEST_FILE_LIST:
@@ -80,7 +84,7 @@ export function createFileGeneratorPanel(context: vscode.ExtensionContext, clien
     currentPanel.webview.html = getWebviewContent(currentPanel.webview, context.extensionUri);
 }
 
-async function preloadFileLists(panel: vscode.WebviewPanel, context: vscode.ExtensionContext) {
+async function preloadFileLists(panel: vscode.WebviewPanel) {
     const mdFiles = await getFileList('md');
     const jsonFiles = await getFileList('json');
 
diff --git a/client/src/copilot/quickPick.ts b/client/src/copilot/quickPick.ts
index d3f6270..51723aa 100644
--- a/client/src/copilot/quickPick.ts
+++ b/client/src/copilot/quickPick.ts
@@ -18,6 +18,7 @@ export function registerQuickPickCommand(context: vscode.ExtensionContext, clien
             { label: GENERAL.QUICK_PICK_OPTION_SETTINGS },
             { label: GENERAL.QUICK_PICK_OPTION_SUGGESTIONS },
 			{ label: GENERAL.QUICK_PICK_OPTION_CHAT_AGENT },
+            { label: GENERAL.QUICK_PICK_OPTION_GENERATOR },
             { label: '', kind: vscode.QuickPickItemKind.Separator },
             { label: enableInlineSuggestions ? GENERAL.QUICK_PICK_OPTION_DISABLE_INLINE_SUGGESTIONS : GENERAL.QUICK_PICK_OPTION_ENABLE_INLINE_SUGGESTIONS },
             { label: enableCodeActions ? GENERAL.QUICK_PICK_OPTION_DISABLE_CODE_ACTIONS : GENERAL.QUICK_PICK_OPTION_ENABLE_CODE_ACTIONS }
@@ -28,8 +29,10 @@ export function registerQuickPickCommand(context: vscode.ExtensionContext, clien
             vscode.commands.executeCommand('cicero-vscode-extension.configureSettings');
         } else if (selection?.label === GENERAL.QUICK_PICK_OPTION_SUGGESTIONS) {
             vscode.commands.executeCommand('cicero-vscode-extension.startPromptProviderUI');
-        } else if (selection?.label === 'Open Chat Agent') {
+        } else if (selection?.label === GENERAL.QUICK_PICK_OPTION_CHAT_AGENT) {
             createOrShowChatPanel(client, context);
+        } else if (selection?.label === GENERAL.QUICK_PICK_OPTION_GENERATOR) {
+            vscode.commands.executeCommand('cicero-vscode-extension.openFileGenerator');
         } else if (selection?.label === GENERAL.QUICK_PICK_OPTION_ENABLE_INLINE_SUGGESTIONS || selection?.label === GENERAL.QUICK_PICK_OPTION_DISABLE_INLINE_SUGGESTIONS) {
             vscode.commands.executeCommand('cicero-vscode-extension.toggleInlineSuggestions');
         } else if (selection?.label === GENERAL.QUICK_PICK_OPTION_ENABLE_CODE_ACTIONS || selection?.label === GENERAL.QUICK_PICK_OPTION_DISABLE_CODE_ACTIONS) {
diff --git a/client/src/copilot/settingsView/configSetting.ts b/client/src/copilot/settingsView/configSetting.ts
index a969146..db4034d 100644
--- a/client/src/copilot/settingsView/configSetting.ts
+++ b/client/src/copilot/settingsView/configSetting.ts
@@ -1,9 +1,10 @@
 import * as vscode from 'vscode';
-import { ASSETS, CONFIG_DEFAULTS } from '../utils/constants';
+import { ASSETS, CONFIG_DEFAULTS, COPILOT_SETTINGS } from '../utils/constants';
 import { htmlTemplate } from './templates/settingsView';
 import { cssTemplate } from './templates/settingsStyle';
 import { scriptTemplate } from './templates/settingScript';
 import { checkCopilotHealth, copilotHealthStatus } from '../healthCheck';
+import { log } from '../../log';
 
 export function createSettingsWebview(context: vscode.ExtensionContext, client: any) {
     const column = vscode.ViewColumn.Beside;
@@ -25,35 +26,60 @@ export function createSettingsWebview(context: vscode.ExtensionContext, client:
     panel.iconPath = iconPath;
 
     const config = vscode.workspace.getConfiguration('cicero-vscode-extension');
-    const configValues = {
-        apiKey: config.get<string>('apiKey', CONFIG_DEFAULTS.apiKey),
-        apiUrl: config.get<string>('apiUrl', CONFIG_DEFAULTS.apiUrl),
-        provider: config.get<string>('provider', CONFIG_DEFAULTS.provider),
-        llmModel: config.get<string>('llmModel', CONFIG_DEFAULTS.llmModel),
-        additionalParams: JSON.stringify(config.get<object>('additionalParams', CONFIG_DEFAULTS.additionalParams), null, 2)
-    };
-
-    panel.webview.html = htmlTemplate(cssTemplate, scriptTemplate, configValues);
 
     panel.webview.onDidReceiveMessage(async message => {
         switch (message.command) {
             case 'saveSettings':
                 const target = message.scope === 'workspace' ? vscode.ConfigurationTarget.Workspace : vscode.ConfigurationTarget.Global;
+
+                if (message.scope === 'workspace' && !vscode.workspace.workspaceFolders) {
+                    vscode.window.showErrorMessage(COPILOT_SETTINGS.CONNECTION_FAILED);
+                    panel.webview.postMessage({ command: 'showError', message: COPILOT_SETTINGS.WORKSPACE_REQUIRED });
+                    return;
+                }
+
+                const sanitizedLLMModel = message.llmModel.trim();
+
                 await config.update('apiKey', message.apiKey, target);
                 await config.update('provider', message.provider, target);
-                await config.update('llmModel', message.llmModel, target);
+                await config.update('llmModel', sanitizedLLMModel, target);
                 await config.update('additionalParams', JSON.parse(message.additionalParams), target);
                 
-                vscode.window.showInformationMessage('Configuration updated successfully!');
+                vscode.window.showInformationMessage(COPILOT_SETTINGS.CONFIG_UPDATED_SUCCESS);
 
                 await checkCopilotHealth(client)
-                if (copilotHealthStatus === true) 
-                    vscode.window.showInformationMessage('Connection to Copilot established successfully!');
-                else
-                    vscode.window.showErrorMessage('Connection to Copilot failed!');
+                log('Copilot settings updated'+copilotHealthStatus);
+                if (copilotHealthStatus === true){
+                    vscode.window.showInformationMessage(COPILOT_SETTINGS.CONNECTION_SUCCESS);
+                    panel.webview.postMessage({ command: 'hideError' });
+                }
+                else {
+                    vscode.window.showErrorMessage(COPILOT_SETTINGS.CONNECTION_FAILED);
+                    panel.webview.postMessage({ command: 'showError', message: COPILOT_SETTINGS.CONNECTION_FAILED_MESSAGE });
+                }
                 
                 break;
         }
     });
+
+    panel.onDidChangeViewState(() => {
+        if (panel.visible) {
+            let config = vscode.workspace.getConfiguration('cicero-vscode-extension');
+            const updatedConfigValues = {
+                apiKey: config.get<string>('apiKey', CONFIG_DEFAULTS.apiKey),
+                apiUrl: config.get<string>('apiUrl', CONFIG_DEFAULTS.apiUrl),
+                provider: config.get<string>('provider', CONFIG_DEFAULTS.provider),
+                llmModel: config.get<string>('llmModel', CONFIG_DEFAULTS.llmModel),
+                additionalParams: JSON.stringify(config.get<object>('additionalParams', CONFIG_DEFAULTS.additionalParams), null, 2)
+            };
+            panel.webview.html = htmlTemplate(cssTemplate, scriptTemplate, updatedConfigValues);
+        }
+    });
+
+    panel.onDidDispose(() => {
+        panel.dispose();
+    });
+
+    panel.options
 }
 
diff --git a/client/src/copilot/settingsView/templates/settingScript.ts b/client/src/copilot/settingsView/templates/settingScript.ts
index 1e8a4a6..38623b5 100644
--- a/client/src/copilot/settingsView/templates/settingScript.ts
+++ b/client/src/copilot/settingsView/templates/settingScript.ts
@@ -53,4 +53,37 @@ export const scriptTemplate = `
         }
     }
     
+    function updateLLMModel() {
+        const provider = document.getElementById('provider').value;
+        const llmModelInput = document.getElementById('llmModel');
+        let defaultModel = '';
+
+        switch (provider) {
+            case 'gemini':
+                defaultModel = 'gemini-pro';
+                break;
+            case 'openai':
+                defaultModel = 'gpt-3.5-turbo';
+                break;
+            case 'mistral':
+                defaultModel = 'mistral-large-latest';
+                break;
+        }
+
+        llmModelInput.value = defaultModel;
+    }
+
+    window.addEventListener('message', event => {
+        const message = event.data;
+        const errorElement = document.getElementById('copilotHealthError');
+        switch (message.command) {
+            case 'showError':
+                errorElement.innerText = message.message;
+                errorElement.style.display = 'block';
+                break;
+            case 'hideError':
+                errorElement.style.display = 'none';
+                break;
+        }
+    });
 `;
diff --git a/client/src/copilot/settingsView/templates/settingsStyle.ts b/client/src/copilot/settingsView/templates/settingsStyle.ts
index 1e79a0f..23033ba 100644
--- a/client/src/copilot/settingsView/templates/settingsStyle.ts
+++ b/client/src/copilot/settingsView/templates/settingsStyle.ts
@@ -117,7 +117,7 @@ export const cssTemplate = `
         color: var(--input-foreground); 
     }
     .error-message { /* New class for error messages */
-        color: var(--error-foreground);
+        color: var(--vscode-list-errorForeground);
         margin-top: 5px;
         font-size: 0.9em;
         display: none; /* Hidden by default */
diff --git a/client/src/copilot/settingsView/templates/settingsView.ts b/client/src/copilot/settingsView/templates/settingsView.ts
index 8a98829..c9ac9c8 100644
--- a/client/src/copilot/settingsView/templates/settingsView.ts
+++ b/client/src/copilot/settingsView/templates/settingsView.ts
@@ -29,7 +29,7 @@ export const htmlTemplate = (css: string, script: string, configValues: { [key:
                 <label for="provider">Provider <span class="required-asterisk">*</span></label>
                 <p>Select the AI provider.</p>
                 <div class="select-container">
-                    <select id="provider" style="width: 100%;">
+                    <select id="provider" style="width: 100%;" onchange="updateLLMModel()">
                         <option value="gemini" ${configValues.provider === 'gemini' ? 'selected' : ''}>Gemini</option>
                         <option value="openai" ${configValues.provider === 'openai' ? 'selected' : ''}>OpenAI</option>
                         <option value="mistral" ${configValues.provider === 'mistral' ? 'selected' : ''}>MistralAI</option>
@@ -39,7 +39,7 @@ export const htmlTemplate = (css: string, script: string, configValues: { [key:
                 <span class="error-message" id="providerError"></span>         
             </div>
             <div class="form-group">
-                <label for="llmModel">LLM Model</label>
+                <label for="llmModel">LLM Model <span class="required-asterisk">*</span></label>
                 <p>The specific language model (default: Gemini - gemini-pro, OpenAI - gpt-3.5-turbo, MistralAI - mistral-large-latest).</p>
                 <input type="text" id="llmModel" value="${configValues.llmModel}">
                 <span class="error-message" id="llmModelError"></span>
@@ -54,13 +54,14 @@ export const htmlTemplate = (css: string, script: string, configValues: { [key:
                 <p>Select whether to save the settings globally (for all projects) or for the current workspace only.</p>
                 <div class="select-container">
                     <select id="scope" style="width: 100%;">
-                        <option value="workspace" selected>Workspace</option>
-                        <option value="global">Global</option>
+                        <option value="workspace">Workspace</option>
+                        <option value="global" selected>Global</option>
                     </select>
                     <i class="codicon codicon-chevron-down"></i>
                 </div>
             </div>
             <button class="button" onclick="saveSettings()">Save Settings</button>
+            <span class="error-message" id="copilotHealthError"></span>
         </div>
         <script>
             ${script}
diff --git a/client/src/copilot/utils/constants.ts b/client/src/copilot/utils/constants.ts
index a2195bc..5fe9285 100644
--- a/client/src/copilot/utils/constants.ts
+++ b/client/src/copilot/utils/constants.ts
@@ -19,7 +19,8 @@ export const GENERAL = {
     QUICK_PICK_STATUS_OK: '$(copilot)  Status:  Ready',
     QUICK_PICK_STATUS_ERROR: '$(copilot)  Status:  Configuration Error',
     QUICK_PICK_PLACEHOLDER: 'Choose an action',
-    QUICK_PICK_OPTION_CHAT_AGENT: 'Open Chat Agent',
+    QUICK_PICK_OPTION_CHAT_AGENT: '$(comment-discussion)  Open Chat Agent',
+    QUICK_PICK_OPTION_GENERATOR: '$(wand)  Model Generation Wizard',
     QUICK_PICK_OPTION_SETTINGS: '$(gear)  Open Copilot Settings',
     QUICK_PICK_OPTION_SUGGESTIONS: '$(copilot)  Get Inline Suggestions',
     QUICK_PICK_OPTION_ENABLE_INLINE_SUGGESTIONS: 'Enable Inline Suggestions',
@@ -87,7 +88,7 @@ export const DEFAULT_LLM_MODELS = {
 export const DEFAULT_LLM_ENDPOINTS = {
     GEMINI: 'https://generativelanguage.googleapis.com/v1beta/models/gemini-pro:generateContent',
     OPENAI: 'https://api.openai.com/v1/chat/completions',
-    MISTRALAI: 'https://api.mistralai.com/v1/chat/completions'
+    MISTRALAI: 'https://api.mistral.ai/v1/chat/completions'
 };
   
 // Constants for Panel Manager
@@ -120,4 +121,11 @@ export const ASSETS = {
     ACCORD_LOGO_DARK: 'assets/dark/accord_logo.png',
     ACCORD_LOGO_LIGHT: 'assets/light/accord_logo.png'
 };
-  
\ No newline at end of file
+
+export const COPILOT_SETTINGS = {
+    CONFIG_UPDATED_SUCCESS: 'Configuration updated successfully!',
+    CONNECTION_SUCCESS: 'Connection to Copilot established successfully!',
+    CONNECTION_FAILED: 'Connection to Copilot failed!',
+    CONNECTION_FAILED_MESSAGE: 'Connection to Copilot failed! Please check your API key, billing status, and LLM model.',
+    WORKSPACE_REQUIRED: 'Cannot save settings with "Workspace" scope because no workspace is open. Please save them as Global settings.'
+};
\ No newline at end of file
diff --git a/package.json b/package.json
index 47dab20..e8c6170 100644
--- a/package.json
+++ b/package.json
@@ -167,28 +167,6 @@
         "title": "Open File Generator"
       }
     ],
-    "viewsContainers": {
-      "activitybar": [
-        {
-          "id": "chat",
-          "title": "Chat",
-          "icon": "comments-view-icon"
-        },
-        {
-          "id": "fileGenerator",
-          "title": "File Generator",
-          "icon": "file-generator-view-icon"
-        }
-      ]
-    },
-    "views": {
-      "chat": [
-        {
-          "id": "chatPanel",
-          "name": "Accord Assistant"
-        }
-      ]
-    },
     "submenus": [
       {
         "id": "accordProjectSubmenu",
@@ -239,9 +217,6 @@
         },
         {
           "command": "cicero-vscode-extension.configureSettings"
-        },
-        {
-          "command": "cicero-vscode-extension.openFileGenerator"
         }
       ]
     }
diff --git a/server/src/commands/commandHandler.ts b/server/src/commands/commandHandler.ts
index 3cf236f..ba8a128 100644
--- a/server/src/commands/commandHandler.ts
+++ b/server/src/commands/commandHandler.ts
@@ -69,7 +69,7 @@ export async function loadProjectFiles(fileExtensions:string[]) {
 		log(`Document count: ${GLOBAL_STATE.documents.all().length}`);
 		GLOBAL_STATE.documents.all().forEach(async document => {
 			const change: TextDocumentChangeEvent<TextDocument> = { document };
-			handleConcertoDocumentChange(GLOBAL_STATE, change);
+			await handleConcertoDocumentChange(GLOBAL_STATE, change);
 		});
 	} else {
 		log('GLOBAL_STATE.connection is null');
diff --git a/server/src/copilot/agent/promptTemplates/concertoInlineTemplate.ts b/server/src/copilot/agent/promptTemplates/concertoInlineTemplate.ts
index 80893c3..7248092 100644
--- a/server/src/copilot/agent/promptTemplates/concertoInlineTemplate.ts
+++ b/server/src/copilot/agent/promptTemplates/concertoInlineTemplate.ts
@@ -13,7 +13,7 @@ export function getConcertoInlineTemplate(beforeCursor: string, afterCursor: str
             role: "user"
         },
         {
-            content: `${beforeCursor} /* Analyze the following code in concerto and complete the code based on the context. ${instruction}. Remove this commented instruction in complete code. */ ${afterCursor}`,
+            content: `${beforeCursor} /* The user's cursor is here. Analyze the following code in concerto language and provide a suggestion to complete it according to this instruction from user: ${instruction}. Only add new code at the cursor position without modifying the existing code. Remove this comment in the final output. */ ${afterCursor}. Start output from here ${beforeCursor}`,
             role: "user"
         }
     ];
diff --git a/server/src/copilot/agent/promptTemplates/inlineTemplate.ts b/server/src/copilot/agent/promptTemplates/inlineTemplate.ts
index 3915c27..2d645a4 100644
--- a/server/src/copilot/agent/promptTemplates/inlineTemplate.ts
+++ b/server/src/copilot/agent/promptTemplates/inlineTemplate.ts
@@ -9,7 +9,7 @@ export function getInlineTemplate(beforeCursor: string, afterCursor: string, pro
             role: "system"
         },
         {
-            content: `${beforeCursor} /* Analyze the following code in ${language} and complete the code based on the context. ${instruction}. Remove this commented instruction in complete code. */ ${afterCursor}`,
+            content: `${beforeCursor} /* The user's cursor is here. Analyze the following code in ${language} language and complete it according to this instruction from user: ${instruction}. Only add new code at the cursor position without modifying the existing code. Remove this comment in the final output. */ ${afterCursor}. Start output from here ${beforeCursor}`,
             role: "user"
         }
     ];
diff --git a/server/src/copilot/llm/llmManager.ts b/server/src/copilot/llm/llmManager.ts
index 3413ebe..66f3736 100644
--- a/server/src/copilot/llm/llmManager.ts
+++ b/server/src/copilot/llm/llmManager.ts
@@ -41,6 +41,11 @@ async function handleErrors(updatedContent: string, promptConfig: PromptConfig,
         promptConfig.previousError = errors;
     }
 
+    if (GLOBAL_STATE.connection) {
+        GLOBAL_STATE.diagnostics.clearErrors(tempDocumentName, 'model');
+        GLOBAL_STATE.diagnostics.send(GLOBAL_STATE.connection);
+    }
+
     return errors;
 }
 
@@ -50,10 +55,11 @@ export async function generateContent(config: ModelConfig, documents: Documents,
         let shouldCache = false;
         let errors: any[] = [];
         let iteration = 0;
-        log('Generating content for document: ' + documents);
         const documentDetails: DocumentDetails = documents.main;
 
-        const cacheKey = generateCacheKey(documentDetails, promptConfig, config);
+        log('Generating content for document :' + (documentDetails.fileName || ''));
+
+        const cacheKey = generateCacheKey(documents, promptConfig, config);
         const maxRetries = DEFAULTS.MAX_RETRIES;
 
         try {
@@ -65,9 +71,9 @@ export async function generateContent(config: ModelConfig, documents: Documents,
                 else
                     log('Fixing errors in generated content from model:' + provider + ' Attempt: ' + iteration);    
                 
-                const promptArray = await agentPlanner({ documents, promptConfig, config});
                 const cachedResponse = getPromptFromCache(cacheKey);
                 if (!cachedResponse) {
+                    const promptArray = await agentPlanner({ documents, promptConfig, config});
                     generatedContent = await generateContentByProvider(provider, config, promptArray);
                     if (promptConfig.requestType === 'inline') {
                         const documentContent = documentDetails.content;
diff --git a/server/src/copilot/llm/providers/gemini.ts b/server/src/copilot/llm/providers/gemini.ts
index 5813da0..ef7f9aa 100644
--- a/server/src/copilot/llm/providers/gemini.ts
+++ b/server/src/copilot/llm/providers/gemini.ts
@@ -51,12 +51,9 @@ class Gemini implements LargeLanguageModel {
     }
 
     async generateEmbeddings(config: any, text: string): Promise<Embedding[]> {
-        let { apiUrl, embeddingModel } = config;
+        let { embeddingModel } = config;
         const { accessToken } = config;
-
-        if (!apiUrl) {
-            apiUrl = GEMINI_ENDPOINTS.EMBEDDINGS;
-        }
+        let apiUrl = GEMINI_ENDPOINTS.EMBEDDINGS;
 
         embeddingModel = embeddingModel || GEMINI_ENDPOINTS.EMBEDDING_MODEL;
         const updatedApiUrl = this.replaceModelInApiUrl(apiUrl, embeddingModel);
diff --git a/server/src/copilot/llm/providers/mistral.ts b/server/src/copilot/llm/providers/mistral.ts
index 43b30c8..4dd55f3 100644
--- a/server/src/copilot/llm/providers/mistral.ts
+++ b/server/src/copilot/llm/providers/mistral.ts
@@ -50,15 +50,11 @@ class Mistral implements LargeLanguageModel {
     }
 
     async generateEmbeddings(config: any, text: string): Promise<Embedding[]> {
-        let { apiUrl, embeddingModel } = config;
-        const { accessToken } = config;
-
-        if (!apiUrl) {
-            apiUrl = MISTRAL_ENDPOINTS.EMBEDDINGS;
-        }
+        const { embeddingModel, accessToken } = config;
+        let apiUrl = MISTRAL_ENDPOINTS.EMBEDDINGS;
 
-        embeddingModel = embeddingModel || MISTRAL_ENDPOINTS.EMBEDDING_MODEL;
-        const request = this.createGenerateEmbeddingsRequest(text, embeddingModel);
+        let model = embeddingModel || MISTRAL_ENDPOINTS.EMBEDDING_MODEL;
+        const request = this.createGenerateEmbeddingsRequest(text, model);
 
         try {
             const response = await fetch(apiUrl, {
diff --git a/server/src/copilot/llm/providers/openai.ts b/server/src/copilot/llm/providers/openai.ts
index 892d251..6b47670 100644
--- a/server/src/copilot/llm/providers/openai.ts
+++ b/server/src/copilot/llm/providers/openai.ts
@@ -50,15 +50,11 @@ class OpenAI implements LargeLanguageModel {
     }
 
     async generateEmbeddings(config: any, text: string): Promise<Embedding[]> {
-        let { apiUrl, embeddingModel } = config;
-        const { accessToken } = config;
+        const { embeddingModel, accessToken } = config
+        let apiUrl = OPENAI_ENDPOINTS.EMBEDDINGS;
         
-        if (!apiUrl) {
-            apiUrl = OPENAI_ENDPOINTS.EMBEDDINGS;
-        }
-
-        embeddingModel = embeddingModel || OPENAI_ENDPOINTS.EMBEDDING_MODEL;
-        const request = this.createGenerateEmbeddingsRequest(text, embeddingModel);
+        let model = embeddingModel || OPENAI_ENDPOINTS.EMBEDDING_MODEL;
+        const request = this.createGenerateEmbeddingsRequest(text, model);
 
         try {
             const response = await fetch(apiUrl, {
@@ -138,7 +134,7 @@ class OpenAI implements LargeLanguageModel {
 }
 
 export const OPENAI_ENDPOINTS = {
-    CONTENT: 'https://api.openai.com/v1/completions',
+    CONTENT: 'https://api.openai.com/v1/chat/completions',
     EMBEDDINGS: 'https://api.openai.com/v1/embeddings',
     EMBEDDING_MODEL: 'text-embedding-ada-002',
 };
diff --git a/server/src/copilot/utils/cacheKeyGenerator.ts b/server/src/copilot/utils/cacheKeyGenerator.ts
index bd580bd..f965885 100644
--- a/server/src/copilot/utils/cacheKeyGenerator.ts
+++ b/server/src/copilot/utils/cacheKeyGenerator.ts
@@ -1,11 +1,17 @@
-import { DocumentDetails, PromptConfig, ModelConfig } from './types';
+import { PromptConfig, ModelConfig, Documents } from './types';
 
 export function generateCacheKey(
-    documentDetails: DocumentDetails,
+    documents: Documents,
     promptConfig: PromptConfig,
     modelConfig: ModelConfig
 ): string {
-    const { content, cursorPosition } = documentDetails;
+    const { main, contextDocuments } = documents;
+    const { content, cursorPosition } = main;
+    let contextContent = '';
+
+    if (contextDocuments)
+        contextContent = contextDocuments.map((doc) => doc.content).join('');
+
     const { requestType, language, instruction } = promptConfig;
     const { provider, llmModel, apiUrl, additionalParams, accessToken } = modelConfig;
 
@@ -20,5 +26,6 @@ export function generateCacheKey(
         apiUrl || '',
         accessToken,
         JSON.stringify(additionalParams || {}),
+        contextContent,
     ].join(':');
 }
diff --git a/server/src/copilot/utils/responseProcessor.ts b/server/src/copilot/utils/responseProcessor.ts
index 83d8c9c..d9f8a26 100644
--- a/server/src/copilot/utils/responseProcessor.ts
+++ b/server/src/copilot/utils/responseProcessor.ts
@@ -51,8 +51,8 @@ export function cleanSuggestion(documentContent: string, cursorPosition: number,
     const overlapLength = Math.min(beforeCursor.length, cleanSuggestion.length);
 
     for (let i = 0; i < overlapLength; i++) {
-        const beforeCursorSuffix = beforeCursor.slice(beforeCursor.length - i);
-        const suggestionPrefix = cleanSuggestion.slice(0, i);
+        const beforeCursorSuffix = beforeCursor.slice(beforeCursor.length - i).toLowerCase();
+        const suggestionPrefix = cleanSuggestion.slice(0, i).toLowerCase();
 
         if (beforeCursorSuffix === suggestionPrefix) {
             cleanSuggestion = cleanSuggestion.slice(i);
@@ -67,6 +67,9 @@ export function cleanSuggestion(documentContent: string, cursorPosition: number,
         cleanSuggestion = cleanSuggestion.slice(0, afterCursorIndex);
     }
 
+    if(cleanSuggestion.length === 0 && suggestion.length !== 0)
+        cleanSuggestion = suggestion;
+
     return cleanSuggestion;
 }
 
diff --git a/server/src/copilot/utils/robustFetch.ts b/server/src/copilot/utils/robustFetch.ts
index 5d49112..c702135 100644
--- a/server/src/copilot/utils/robustFetch.ts
+++ b/server/src/copilot/utils/robustFetch.ts
@@ -1,6 +1,6 @@
 import { log } from '../../state';
 
-export async function robustFetch(url: string, options: RequestInit, maxRetries = 3, timeout = 15000): Promise<Response> {
+export async function robustFetch(url: string, options: RequestInit, maxRetries = 3, timeout = 30000): Promise<Response> {
     let attempts = 0;
 
     while (attempts < maxRetries) {
diff --git a/server/src/documents/concertoHandler.ts b/server/src/documents/concertoHandler.ts
index dff3f16..163dc78 100644
--- a/server/src/documents/concertoHandler.ts
+++ b/server/src/documents/concertoHandler.ts
@@ -105,8 +105,13 @@ export async function handleConcertoDocumentChange(state: LanguageServerState, c
 			catch (error: any) {
 				if (!state.isLoading) {
 					// we may be offline? Validate without external models
-					state.modelManager.validateModelFiles();
-					state.diagnostics.pushDiagnostic(DiagnosticSeverity.Warning, change.document, error, 'model');
+					if (!navigator.onLine){
+						state.modelManager.validateModelFiles();
+					}
+					// we only report errors for the document that changed, if error has a fileName
+					if(change.document.uri == error.fileName || !error.fileName){
+						state.diagnostics.pushDiagnostic(DiagnosticSeverity.Warning, change.document, error, 'model');
+					}	
 				}
 				else {
 					log(`Ignored model validation error while initializing ${change.document.uri}`);
@@ -114,8 +119,10 @@ export async function handleConcertoDocumentChange(state: LanguageServerState, c
 			}
 		}
 		catch (error: any) {
-			if (!state.isLoading) {
-				state.diagnostics.pushDiagnostic(DiagnosticSeverity.Error, change.document, error, 'model');
+			if(!state.isLoading) {
+				if(change.document.uri == error.fileName || !error.fileName){
+					state.diagnostics.pushDiagnostic(DiagnosticSeverity.Error, change.document, error, 'model');
+				}			
 			}
 			else {
 				log(`Ignored model validation error while initializing ${change.document.uri}`);