Update private_gpt/components/llm/llm_component.py

private_gpt/components/llm/llm_component.py

@@ -60,16 +60,48 @@ class LLMComponent:
             case "mock":
                 self.llm = MockLLM()
 
+
+            case "dynamic":
+                from llama_index.llms import OpenAI
+                openai_settings = settings.openai.api_key
+
+                #default startup
+                logger.info("Initializing the GPT Model in=%s", "gpt-3.5-turbo")
+                self.llm = OpenAI(model="gpt-3.5-turbo", api_key=os.environ.get("OPENAI_API_KEY"))
+
+
     @inject
     def switch_model(self, new_model: str, settings: Settings) -> None:
+        from llama_index.llms import LlamaCPP
         openai_settings = settings.openai.api_key
-        if type(self.llm) == OpenAI:
-            if new_model == "gpt-3.5-turbo":
-                self.llm = OpenAI(model="gpt-3.5-turbo", api_key=os.environ.get("OPENAI_API_KEY"))
-            elif new_model == "gpt-4":
-                # Initialize with the new model 
-                self.llm = OpenAI(model="gpt-4", api_key=os.environ.get("OPENAI_API_KEY"))
-                logger.info("Initializing the GPT Model in=%s", "gpt-4")
+        
+        if new_model == "gpt-3.5-turbo":
+            self.llm = OpenAI(model="gpt-3.5-turbo", api_key=os.environ.get("OPENAI_API_KEY"))
+            
+        elif new_model == "gpt-4":
+            # Initialize with the new model 
+            self.llm = OpenAI(model="gpt-4", api_key=os.environ.get("OPENAI_API_KEY"))
+            logger.info("Initializing the GPT Model in=%s", "gpt-4")
+
+        
+        elif new_model == "mistral-7B":
+            prompt_style_cls = get_prompt_style(settings.local.prompt_style)
+            prompt_style = prompt_style_cls(
+                default_system_prompt=settings.local.default_system_prompt
+            )
+            self.llm = LlamaCPP(
+                model_path=str(models_path / settings.local.llm_hf_model_file),
+                #model_url= "https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF/resolve/main/mistral-7b-instruct-v0.1.Q4_K_M.gguf?download=true",
+                temperature=0.1,
+                max_new_tokens=settings.llm.max_new_tokens,
+                context_window=3900,
+                generate_kwargs={},
+                model_kwargs={"n_gpu_layers": -1},
+                messages_to_prompt=prompt_style.messages_to_prompt,
+                completion_to_prompt=prompt_style.completion_to_prompt,
+                verbose=True,
+            )
+