Update app.py

Update to mixtral-instruct

app.py

@@ -4,7 +4,8 @@ import requests
 from llama_cpp import Llama
 import gradio as gr
 
-url="https://huggingface.co/TheBloke/WizardLM-13B-V1.2-GGUF/resolve/main/wizardlm-13b-v1.2.Q4_0.gguf"
+#url="https://huggingface.co/TheBloke/WizardLM-13B-V1.2-GGUF/resolve/main/wizardlm-13b-v1.2.Q4_0.gguf"
+url="https://huggingface.co/TheBloke/Mixtral-8x7B-Instruct-v0.1-GGUF/resolve/main/mixtral-8x7b-instruct-v0.1.Q4_0.gguf?download=true"
 response = requests.get(url)
 with open("./model.gguf", mode="wb") as file:
   file.write(response.content)
@@ -23,7 +24,9 @@ def response(message, history):
   #url="https://afischer1985-wizardlm-13b-v1-2-q4-0-gguf.hf.space/v1/completions"
   url="http://0.0.0.0:2600/v1/completions"  
   #body={"prompt":"Im Folgenden findest du eine Instruktion, die eine Aufgabe bescheibt. Schreibe eine Antwort, um die Aufgabe zu lösen.\n\n### Instruktion:\n"+message+"\n\n### Antwort:","max_tokens":500, "echo":"False","stream":"True"}
-  body={"prompt":" chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.\n\nUSER:\n"+message+"\n\nASSISTANT:","max_tokens":500, "echo":"False","stream":"True"}   
+  #body={"prompt":" chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.\n\nUSER:\n"+message+"\n\nASSISTANT:","max_tokens":500, "echo":"False","stream":"True"}    
+  #body={"prompt":system+"### Instruktion:\n"+message+"\n\n### Antwort:","max_tokens":500, "echo":"False","stream":"True"} #e.g. SauerkrautLM
+  body={"prompt":"<s>[INST]"+message+"[/INST]### Antwort:","max_tokens":500, "echo":"False","stream":"True"} #e.g. Mixtral-Instruct
   response=""
   buffer=""
   print("URL: "+url)