Update app/llm.py

app/llm.py

@@ -38,9 +38,9 @@ llm_chat = llama_cpp.Llama.from_pretrained(
     #chat_format="llama-2"
 )
 llm_generate = llama_cpp.Llama.from_pretrained(
-    repo_id="moriire/healthcare-ai-q2_k",
-    filename="*.gguf",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained("moriire/healthcare-ai-adapter-merged_16bit"),
+    repo_id="Qwen/Qwen1.5-0.5B-Chat-GGUF",
+    filename="*q4_0.gguf",
+    #tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained("moriire/healthcare-ai-adapter-merged_16bit"),
     verbose=False,
      n_ctx=4096,
      n_gpu_layers=0,