Adding CPU Percentage during generation.

default.env

@@ -1,8 +1,11 @@
 THE_TOKEN = "DISCORD_TOKEN_HERE"
 CHANNEL_IDS = 1094494101631680653,1094628334727614605
 INIT_PROMPT = "Assistant name: ChatAI. You code, write and provide any information without any mistakes."
-# ROOT_IP is only used when running the bot without docker compose
-ROOT_IP = 192.168.0.15
 ROOT_PORT = 8000
 DATA_DIR = /home/USERNAME/weights
 CACHE = 1
+
+# ROOT_IP is only used when running the bot without docker compose
+ROOT_IP = 192.168.0.15
+
+

llamabot.js

@@ -3,7 +3,7 @@ import fetch from 'node-fetch';
 import { emptyResponses } from './assets/emptyMessages.js';
 import { resetResponses, userResetMessages } from './assets/resetMessages.js';
 import { errorMessages, busyResponses } from './assets/errorMessages.js';
-
+import os from 'os';
 import { Client, GatewayIntentBits, ActivityType, Partials } from 'discord.js';
 
 const client = new Client({
@@ -82,7 +82,7 @@ client.on('messageCreate', async (message) => {
 
     // Only respond in the specified channels
     if (!channelIDs.includes(message.channel.id)) {
-    return;
+        return;
     }
 
     if (message.author.bot) return; // Ignore messages from bots
@@ -132,7 +132,7 @@ client.on('messageCreate', async (message) => {
         setPresenceBusy()
         setBusy(message.author.id, true);
 
-        const response = await generateResponse(conversation);
+        const response = await generateResponse(conversation, message);
 
         // Append bot message to conversation history
         conversation.messages.push({
@@ -163,7 +163,8 @@ client.on('messageCreate', async (message) => {
     }
 });
 
-async function generateResponse(conversation) {
+
+async function generateResponse(conversation, message) {
     const controller = new AbortController();
     const timeout = setTimeout(() => {
         controller.abort();
@@ -171,7 +172,26 @@ async function generateResponse(conversation) {
 
     const messagesCopy = [...conversation.messages]; // create a copy of the messages array
 
-    console.log(conversation)
+    let botMessage; // define a variable to hold the message object
+
+    // define a function that shows the system load percentage and updates the message
+    const showSystemLoad = async () => {
+        const systemLoad = os.loadavg()[0] / os.cpus().length * 100;
+        const messageData = `Please wait, I am thinking... System Load: ${systemLoad.toFixed(2)}%`;
+
+        // if the message object doesn't exist, create it
+        if (!botMessage) {
+            botMessage = await message.channel.send(messageData);
+        } else {
+            botMessage.edit(messageData); // otherwise, update the message
+        }
+    };
+
+    // call the function initially
+    await showSystemLoad();
+
+    // refresh the system load percentage and update the message every 5 seconds
+    const refreshInterval = setInterval(showSystemLoad, 5000);
 
     try {
         const response = await fetch(`http://${process.env.ROOT_IP}:${process.env.ROOT_PORT}/v1/chat/completions`, {
@@ -187,10 +207,15 @@ async function generateResponse(conversation) {
         });
 
         const responseData = await response.json();
-        console.log(JSON.stringify(responseData))
+        console.log(JSON.stringify(responseData));
         const choice = responseData.choices[0];
 
-        const responseText = choice.message.content
+        const responseText = choice.message.content;
+
+        // clear the interval, replace the "please wait" message with the response, and update the message
+        clearInterval(refreshInterval);
+        console.log(responseText);
+        botMessage.delete()
 
         return responseText;
 

package.json

@@ -12,6 +12,7 @@
   "dependencies": {
     "discord.js": "^14.9.0",
     "dotenv": "^16.0.3",
-    "node-fetch": "^3.3.1"
+    "node-fetch": "^3.3.1",
+    "os": "^0.1.2"
   }
 }