test: local model

services/model_handler.py

@@ -17,6 +17,30 @@ class LocalHuggingFaceModel(Model):
         self.tokenizer = tokenizer
         self.max_length = max_length
         
+    async def ainvoke(self, prompt: str, **kwargs):
+        """Async invoke method"""
+        return self.invoke(prompt, **kwargs)
+    
+    async def ainvoke_stream(self, prompt: str, **kwargs):
+        """Async streaming invoke method"""
+        yield self.invoke(prompt, **kwargs)
+    
+    def invoke(self, prompt: str, **kwargs):
+        """Synchronous invoke method"""
+        return self.generate(prompt, **kwargs)
+    
+    def invoke_stream(self, prompt: str, **kwargs):
+        """Synchronous streaming invoke method"""
+        yield self.generate(prompt, **kwargs)
+    
+    def parse_provider_response(self, response: str) -> str:
+        """Parse the provider response"""
+        return response
+    
+    def parse_provider_response_delta(self, delta: str) -> str:
+        """Parse the provider response delta for streaming"""
+        return delta
+        
     def generate(self, prompt, **kwargs):
         try:
             inputs = self.tokenizer(prompt, return_tensors="pt", padding=True)