Temporary agents and chats

backend/src/agents/expert.py

@@ -51,48 +51,33 @@ class ExpertAgent(Agent):
         if prompt:
             messages.append(LlamaMessage("user", prompt))
 
-        model = self.config['model']
+        model_id = self.config['model']
+        db_model = await self.app.get_model(model_id)
+        model_engine = db_model.value["engine"]
 
-        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=opts,
+            )
+        elif model_engine == "replicate":
+            if 'parameters' in db_model:
+                opts['parameters'] = db_model['parameters']
 
-            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=opts,
-                )
+            response = await inference.replicate_completion(self.app, messages, model=db_model.value['model'], opts=opts)
 
-                return response
-            elif model_engine == "replicate":
-                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))
-
-        elif model.startswith("ollama:"):
-            response = await inference.ollama_chat_completion(self.app, messages, model=model[7:])
-        elif model == "qwen-big":
-            response = await inference.replicate_qwen_completion(self.app, messages, model=inference.Models.qwen3_235b)
-        elif model == "qwen-small":
-            response = await inference.replicate_qwen_completion(self.app, messages, model=inference.Models.qwen3_8b)
-        elif model == "llama-big":
-            response = await inference.replicate_llama_completion(self.app, messages, model=inference.Models.llama_70b)
-        elif model == "llama-small":
-            response = await inference.replicate_llama_completion(self.app, messages, model=inference.Models.llama_8b)
+        elif model_engine == "ollama":
+            response = await inference.ollama_chat_completion(self.app, messages, model=db_model.value['model'], opts=opts)
         else:
-            # An error here is questionable, model is user configurable
-            raise Exception("Unkonwn agent model: " + model)
+            raise Exception("Unsuported model: " + repr(db_model.value))
 
         return response
 
@@ -108,6 +93,8 @@ class ExpertAgent(Agent):
                 return 'assistant'
             elif sender == 'user':
                 return 'user'
+            elif sender.startswith("tmp-"):
+                return 'assistant'
             else:
                 raise Exception("Unsupported sender: " + sender)