Update src/models.py

src/models.py

@@ -11,8 +11,8 @@ def download_models():
     1. BioMistral-7B - A medical-specific language model.
     2. Llama-2-7B - A general-purpose language model.
     """
-    os.system('wget -O BioMistral-7B.Q4_K_M.gguf "https://huggingface.co/MaziyarPanahi/BioMistral-7B-GGUF/resolve/main/BioMistral-7B.Q4_K_M.gguf?download=true"')
-    os.system('wget -O llama-2-7b.Q4_K_M.gguf "https://huggingface.co/TheBloke/Llama-2-7B-GGUF/resolve/main/llama-2-7b.Q4_K_M.gguf?download=true"')
+    os.system('wget -O llama-2-7b.Q8_0.gguf "https://huggingface.co/TheBloke/Llama-2-7B-GGUF/resolve/main/llama-2-7b.Q8_0.gguf?download=true')
+    os.system("wget -O BioMistral-7B.Q8_0.gguf "https://huggingface.co/MaziyarPanahi/BioMistral-7B-GGUF/resolve/main/BioMistral-7B.Q8_0.gguf?download=true")
 
 # Call the download function to ensure the models are available before starting the bot
 download_models()
@@ -25,14 +25,14 @@ classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnl
 
 # Load LLM models using CTransformers
 general_llm = CTransformers(
-    model="/kaggle/working/llama-2-7b.Q8_0.gguf", 
+    model="./llama-2-7b.Q8_0.gguf", 
     model_type="llama",
     config={'max_new_tokens': 512, 'temperature': 0.7},
     stream=True  # Enable streaming here
 )
 
 medical_llm = CTransformers(
-    model="/kaggle/working/BioMistral-7B.Q8_0.gguf", 
+    model="./BioMistral-7B.Q8_0.gguf", 
     model_type="llama",
     config={'max_new_tokens': 512, 'temperature': 0.7},
     stream=True  # Enable streaming here