Update app.py

app.py

@@ -8,8 +8,8 @@ from huggingface_hub import hf_hub_download
 
 llm = Llama(
     model_path=hf_hub_download(
-        repo_id="galatolo/cerbero-7b-gguf",
-        filename="ggml-model-Q4_K.gguf",
+        repo_id="FinancialSupport/saiga-7b-gguf",
+        filename="saiga-7b.Q4_K_M.gguf",
     ),
     n_ctx=4086,
 ) 
@@ -18,7 +18,7 @@ history = []
 
 def generate_text(message, history):
     temp = ""
-    input_prompt = "Conversazione tra umano ed un assistente AI di nome cerbero-7b\n"
+    input_prompt = "Conversazione tra umano ed un assistente AI di nome saiaga-7b\n"
     for interaction in history:
         input_prompt += "[|Umano|] " + interaction[0] + "\n"
         input_prompt += "[|Assistente|]" + interaction[1]
@@ -50,8 +50,8 @@ def generate_text(message, history):
 
 demo = gr.ChatInterface(
     generate_text,
-    title="cerbero-7b running on CPU (quantized Q4_K)",
-    description="This is a quantized version of cerbero-7b running on CPU (very slow). It is less powerful than the original version, but it can even run on a Raspberry Pi 4.",
+    title="saiga-7b running on CPU (quantized Q4_K)",
+    description="This is a quantized version of saiga-7b running on CPU (very slow). It is less powerful than the original version, but it can even run on the free tier of huggingface.",
     examples=[
         "Dammi 3 idee di ricette che posso fare con i pistacchi",
         "Prepara un piano di esercizi da poter fare a casa",