Reimplement replicate inference. Allow model deletion. Implement repetition penalty. Fixes in stream reading.

backend/src/agents/expert.py

@@ -21,6 +21,7 @@ class ExpertAgent(Agent):
         AgentField('system_prompt', 'string', default="You are an expert AI assistant."),
         AgentField('model', 'model_id', default=None),
         AgentField('temperature', 'number', default=None),
+        AgentField('repetition_penalty', 'number', default=None),
     ]
 
     def __init__(self, app, data):
@@ -55,18 +56,27 @@ class ExpertAgent(Agent):
         if type(model) is int:
             db_model = await self.app.get_model(model)
             model_engine = db_model.value["engine"]
+
+            opts = {}
+            if 'temperature' in self.config:
+                opts['temperature'] = self.config['temperature']
+            if 'repetition_penalty' in self.config:
+                opts['repetition_penalty'] = self.config['repetition_penalty']
             
             if model_engine == "llama-cpp":
                 response = await inference.local_chat_completion(
                     self.app,
                     messages,
                     model=db_model.value['model'],
-                    opts={
-                        'temperature':self.config.get('temperature')
-                    }
+                    opts=opts,
                 )
+
+                return response
             elif model_engine == "replicate":
-                response = await inference.replicate_llama_completion(self.app, messages, model=db_model.value['model'])
+                if 'parameters' in db_model:
+                    opts['parameters'] = db_model['parameters']
+
+                response = await inference.replicate_completion(self.app, messages, model=db_model.value['model'], opts=opts)
             else:
                 raise Exception("Unsuported model: " + repr(db_model.value))