Load model directly from huggingface

main.py

@@ -9,9 +9,10 @@ from typing import List, Dict
 from fastapi.middleware.cors import CORSMiddleware
 
 #Model loading
-model = AutoModelForCausalLM.from_pretrained("vigostral-7b-chat.Q6_K.gguf",
-    model_type='llama',
-    threads = 3,
+model = AutoModelForCausalLM.from_pretrained("TheBloke/Vigostral-7B-Chat-GGUF",
+    model_file="vigostral-7b-chat.Q4_K_M.gguf",
+    model_type="mistral",
+    threads = 3
 )
 
 #Fast API