diff --git a/web/containers/Providers/ModelHandler.tsx b/web/containers/Providers/ModelHandler.tsx
index ad7f5c974..3a6c19e0d 100644
--- a/web/containers/Providers/ModelHandler.tsx
+++ b/web/containers/Providers/ModelHandler.tsx
@@ -302,15 +302,10 @@ export default function ModelHandler() {
 
   const generateThreadTitle = (message: ThreadMessage, thread: Thread) => {
     // If this is the first ever prompt in the thread
-    if (
-      (thread.title ?? thread.metadata?.title)?.trim() !== defaultThreadTitle
-    ) {
+    if ((thread.title ?? thread.metadata?.title)?.trim() !== defaultThreadTitle)
       return
-    }
 
-    if (!activeModelRef.current) {
-      return
-    }
+    if (!activeModelRef.current) return
 
     // Check model engine; we don't want to generate a title when it's not a local engine. remote model using first promp
     if (!isLocalEngine(activeModelRef.current?.engine as InferenceEngine)) {
@@ -332,6 +327,7 @@ export default function ModelHandler() {
             ...updatedThread,
           })
         })
+        .catch(console.error)
     }
 
     // This is the first time message comes in on a new thread
diff --git a/web/hooks/useDeleteThread.ts b/web/hooks/useDeleteThread.ts
index 7b98a4ea5..f69ccd47e 100644
--- a/web/hooks/useDeleteThread.ts
+++ b/web/hooks/useDeleteThread.ts
@@ -37,9 +37,16 @@ export default function useDeleteThread() {
     async (threadId: string) => {
       const thread = threads.find((c) => c.id === threadId)
       if (!thread) return
+      const availableThreads = threads.filter((c) => c.id !== threadId)
+      setThreads(availableThreads)
+
+      // delete the thread state
+      deleteThreadState(threadId)
+
       const assistantInfo = await extensionManager
         .get<ConversationalExtension>(ExtensionTypeEnum.Conversational)
         ?.getThreadAssistant(thread.id)
+        .catch(console.error)
 
       if (!assistantInfo) return
       const model = models.find((c) => c.id === assistantInfo?.model?.id)