adding N_THREADS and MAX_TOKENS vars

default.env

@@ -4,6 +4,8 @@ INIT_PROMPT = "Assistant name: ChatAI. You code, write and provide any informati
 ROOT_PORT = 8000
 DATA_DIR = /home/USERNAME/weights
 CACHE = 1
+N_THREADS = 4
+MAX_TOKENS = 1024
 
 # ROOT_IP is only used when running the bot without docker compose
 ROOT_IP = 192.168.0.15

docker-compose.yml

@@ -24,4 +24,4 @@ services:
       - CHANNEL_IDS
       - ROOT_IP=llama-python-server
       - ROOT_PORT=8000
-      - INIT_PROMPT='Assistant name':' ChatAI. You code, write and provide any information without any mistakes.'
+      - INIT_PROMPT='Assistant Name':' ChatAI. You code, write and provide any information without any mistakes.'

llamabot.js

@@ -231,14 +231,15 @@ async function generateResponse(conversation, message) {
         const response = await fetch(`http://${process.env.ROOT_IP}:${process.env.ROOT_PORT}/v1/chat/completions`, {
             method: 'POST',
             headers: {
-                'accept': 'application/json',
-                'Content-Type': 'application/json'
+              'accept': 'application/json',
+              'Content-Type': 'application/json'
             },
             body: JSON.stringify({
-                messages: messagesCopy // use the copy of the messages array
+              messages: messagesCopy,
+              max_tokens: process.env.MAX_TOKENS // add the max_tokens parameter here
             }),
             signal: controller.signal
-        });
+          });
 
         const responseData = await response.json();
         console.log(JSON.stringify(responseData));