Update app.py

app.py

@@ -1,4 +1,18 @@
+import gradio as gr
 from gpt4all import GPT4All
 from huggingface_hub import hf_hub_download
 
-title = "Mistral-7B-Instruct-GGUF-HOTRSS Run On CPU-Basic Free Hardware"
+title = "Mistral-7B-Instruct-GGUF-HOTRSS Run On CPU-Basic Free Hardware"
+
+max_new_tokens = 2048
+
+def generater(message, history, temperature, top_p, top_k):
+    prompt = "<s>"
+    for user_message, assistant_message in history:
+        prompt += model.config["promptTemplate"].format(user_message)
+        prompt += assistant_message + "</s>"
+    prompt += model.config["promptTemplate"].format(message)
+    outputs = []    
+    for token in model.generate(prompt=prompt, temp=temperature, top_k = top_k, top_p = top_p, max_tokens = max_new_tokens, streaming=True):
+        outputs.append(token)
+        yield "".join(outputs)