diff --git a/gpt4all-chat/CHANGELOG.md b/gpt4all-chat/CHANGELOG.md
index 82ded06a..ddd909b9 100644
--- a/gpt4all-chat/CHANGELOG.md
+++ b/gpt4all-chat/CHANGELOG.md
@@ -13,6 +13,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 ### Changed
 - Rebase llama.cpp on latest upstream as of September 26th ([#2998](https://github.com/nomic-ai/gpt4all/pull/2998))
 - Change the error message when a message is too long ([#3004](https://github.com/nomic-ai/gpt4all/pull/3004))
+- Simplify chatmodel to get rid of unnecessary field and bump chat version ([#3016](https://github.com/nomic-ai/gpt4all/pull/3016))
 
 ### Fixed
 - Fix a crash when attempting to continue a chat loaded from disk ([#2995](https://github.com/nomic-ai/gpt4all/pull/2995))
diff --git a/gpt4all-chat/qml/ChatView.qml b/gpt4all-chat/qml/ChatView.qml
index f62b6654..3bc77219 100644
--- a/gpt4all-chat/qml/ChatView.qml
+++ b/gpt4all-chat/qml/ChatView.qml
@@ -1694,19 +1694,21 @@ Rectangle {
                         imageHeight: 20
                         visible: chatModel.count && !currentChat.isServer && currentChat.isModelLoaded && !currentChat.responseInProgress
                         onClicked: {
-                            var index = Math.max(0, chatModel.count - 1);
-                            var listElement = chatModel.get(index);
+                            if (chatModel.count < 2)
+                                return
+                            var promptIndex = chatModel.count - 2
+                            var promptElement = chatModel.get(promptIndex)
+                            var responseIndex = chatModel.count - 1
+                            var responseElement = chatModel.get(responseIndex)
+                            if (promptElement.name !== "Prompt: " || responseElement.name !== "Response: ")
+                                return
                             currentChat.regenerateResponse()
-                            if (chatModel.count) {
-                                if (listElement.name === "Response: ") {
-                                    chatModel.updateCurrentResponse(index, true);
-                                    chatModel.updateStopped(index, false);
-                                    chatModel.updateThumbsUpState(index, false);
-                                    chatModel.updateThumbsDownState(index, false);
-                                    chatModel.updateNewResponse(index, "");
-                                    currentChat.prompt(listElement.prompt)
-                                }
-                            }
+                            chatModel.updateCurrentResponse(responseIndex, true)
+                            chatModel.updateStopped(responseIndex, false)
+                            chatModel.updateThumbsUpState(responseIndex, false)
+                            chatModel.updateThumbsDownState(responseIndex, false)
+                            chatModel.updateNewResponse(responseIndex, "")
+                            currentChat.prompt(promptElement.value)
                         }
                         ToolTip.visible: regenerateButton.hovered
                         ToolTip.text: qsTr("Redo last chat response")
diff --git a/gpt4all-chat/src/chat.cpp b/gpt4all-chat/src/chat.cpp
index 3a313887..6ee7b30f 100644
--- a/gpt4all-chat/src/chat.cpp
+++ b/gpt4all-chat/src/chat.cpp
@@ -240,7 +240,7 @@ void Chat::newPromptResponsePair(const QString &prompt)
     m_chatModel->updateCurrentResponse(m_chatModel->count() - 1, false);
     // the prompt is passed as the prompt item's value and the response item's prompt
     m_chatModel->appendPrompt("Prompt: ", prompt);
-    m_chatModel->appendResponse("Response: ", prompt);
+    m_chatModel->appendResponse("Response: ");
     emit resetResponseRequested();
 }
 
@@ -251,7 +251,7 @@ void Chat::serverNewPromptResponsePair(const QString &prompt)
     m_chatModel->updateCurrentResponse(m_chatModel->count() - 1, false);
     // the prompt is passed as the prompt item's value and the response item's prompt
     m_chatModel->appendPrompt("Prompt: ", prompt);
-    m_chatModel->appendResponse("Response: ", prompt);
+    m_chatModel->appendResponse("Response: ");
 }
 
 bool Chat::restoringFromText() const
diff --git a/gpt4all-chat/src/chatlistmodel.cpp b/gpt4all-chat/src/chatlistmodel.cpp
index b4afb39f..c5be4338 100644
--- a/gpt4all-chat/src/chatlistmodel.cpp
+++ b/gpt4all-chat/src/chatlistmodel.cpp
@@ -19,7 +19,7 @@
 #include <algorithm>
 
 #define CHAT_FORMAT_MAGIC 0xF5D553CC
-#define CHAT_FORMAT_VERSION 9
+#define CHAT_FORMAT_VERSION 10
 
 class MyChatListModel: public ChatListModel { };
 Q_GLOBAL_STATIC(MyChatListModel, chatListModelInstance)
diff --git a/gpt4all-chat/src/chatmodel.h b/gpt4all-chat/src/chatmodel.h
index f061ccf7..2b441a79 100644
--- a/gpt4all-chat/src/chatmodel.h
+++ b/gpt4all-chat/src/chatmodel.h
@@ -22,7 +22,6 @@ struct ChatItem
     Q_PROPERTY(int id MEMBER id)
     Q_PROPERTY(QString name MEMBER name)
     Q_PROPERTY(QString value MEMBER value)
-    Q_PROPERTY(QString prompt MEMBER prompt)
     Q_PROPERTY(QString newResponse MEMBER newResponse)
     Q_PROPERTY(bool currentResponse MEMBER currentResponse)
     Q_PROPERTY(bool stopped MEMBER stopped)
@@ -36,7 +35,6 @@ public:
     int id = 0;
     QString name;
     QString value;
-    QString prompt;
     QString newResponse;
     QList<ResultInfo> sources;
     QList<ResultInfo> consolidatedSources;
@@ -59,7 +57,6 @@ public:
         IdRole = Qt::UserRole + 1,
         NameRole,
         ValueRole,
-        PromptRole,
         NewResponseRole,
         CurrentResponseRole,
         StoppedRole,
@@ -88,8 +85,6 @@ public:
                 return item.name;
             case ValueRole:
                 return item.value;
-            case PromptRole:
-                return item.prompt;
             case NewResponseRole:
                 return item.newResponse;
             case CurrentResponseRole:
@@ -115,7 +110,6 @@ public:
         roles[IdRole] = "id";
         roles[NameRole] = "name";
         roles[ValueRole] = "value";
-        roles[PromptRole] = "prompt";
         roles[NewResponseRole] = "newResponse";
         roles[CurrentResponseRole] = "currentResponse";
         roles[StoppedRole] = "stopped";
@@ -137,12 +131,11 @@ public:
         emit countChanged();
     }
 
-    void appendResponse(const QString &name, const QString &prompt)
+    void appendResponse(const QString &name)
     {
         ChatItem item;
         item.id = m_chatItems.count(); // This is only relevant for responses
         item.name = name;
-        item.prompt = prompt;
         item.currentResponse = true;
         beginInsertRows(QModelIndex(), m_chatItems.size(), m_chatItems.size());
         m_chatItems.append(item);
@@ -266,7 +259,6 @@ public:
             stream << c.id;
             stream << c.name;
             stream << c.value;
-            stream << c.prompt;
             stream << c.newResponse;
             stream << c.currentResponse;
             stream << c.stopped;
@@ -336,7 +328,11 @@ public:
             stream >> c.id;
             stream >> c.name;
             stream >> c.value;
-            stream >> c.prompt;
+            if (version < 10) {
+                // This is deprecated and no longer used
+                QString prompt;
+                stream >> prompt;
+            }
             stream >> c.newResponse;
             stream >> c.currentResponse;
             stream >> c.stopped;