feat: system prompt (#448)

embedchain/apps/App.py

@@ -25,6 +25,8 @@ class App(EmbedChain):
 
     def get_llm_model_answer(self, prompt, config: ChatConfig):
         messages = []
+        if config.system_prompt:
+            messages.append({"role": "system", "content": config.system_prompt})
         messages.append({"role": "user", "content": prompt})
         response = openai.ChatCompletion.create(
             model=config.model or "gpt-3.5-turbo-0613",

embedchain/apps/CustomApp.py

@@ -1,5 +1,5 @@
 import logging
-from typing import List
+from typing import List, Optional
 
 from langchain.schema import BaseMessage
 
@@ -84,7 +84,7 @@ class CustomApp(EmbedChain):
         if config.top_p and config.top_p != 1:
             logging.warning("Config option `top_p` is not supported by this model.")
 
-        messages = CustomApp._get_messages(prompt)
+        messages = CustomApp._get_messages(prompt, system_prompt=config.system_prompt)
 
         return chat(messages).content
 
@@ -97,7 +97,7 @@ class CustomApp(EmbedChain):
         if config.max_tokens and config.max_tokens != 1000:
             logging.warning("Config option `max_tokens` is not supported by this model.")
 
-        messages = CustomApp._get_messages(prompt)
+        messages = CustomApp._get_messages(prompt, system_prompt=config.system_prompt)
 
         return chat(messages).content
 
@@ -110,7 +110,7 @@ class CustomApp(EmbedChain):
         if config.top_p and config.top_p != 1:
             logging.warning("Config option `top_p` is not supported by this model.")
 
-        messages = CustomApp._get_messages(prompt)
+        messages = CustomApp._get_messages(prompt, system_prompt=config.system_prompt)
 
         return chat(messages).content
 
@@ -133,15 +133,19 @@ class CustomApp(EmbedChain):
         if config.top_p and config.top_p != 1:
             logging.warning("Config option `top_p` is not supported by this model.")
 
-        messages = CustomApp._get_messages(prompt)
+        messages = CustomApp._get_messages(prompt, system_prompt=config.system_prompt)
 
         return chat(messages).content
 
     @staticmethod
-    def _get_messages(prompt: str) -> List[BaseMessage]:
+    def _get_messages(prompt: str, system_prompt: Optional[str] = None) -> List[BaseMessage]:
         from langchain.schema import HumanMessage, SystemMessage
 
-        return [SystemMessage(content="You are a helpful assistant."), HumanMessage(content=prompt)]
+        messages = []
+        if system_prompt:
+            messages.append(SystemMessage(content=system_prompt))
+        messages.append(HumanMessage(content=prompt))
+        return messages
 
     def _stream_llm_model_response(self, response):
         """

embedchain/apps/Llama2App.py

@@ -2,7 +2,7 @@ import os
 
 from langchain.llms import Replicate
 
-from embedchain.config import AppConfig
+from embedchain.config import AppConfig, ChatConfig
 from embedchain.embedchain import EmbedChain
 
 
@@ -27,8 +27,10 @@ class Llama2App(EmbedChain):
 
         super().__init__(config)
 
-    def get_llm_model_answer(self, prompt, config: AppConfig = None):
+    def get_llm_model_answer(self, prompt, config: ChatConfig = None):
         # TODO: Move the model and other inputs into config
+        if config.system_prompt:
+            raise ValueError("Llama2App does not support `system_prompt`")
         llm = Replicate(
             model="a16z-infra/llama13b-v2-chat:df7690f1994d94e96ad9d568eac121aecf50684a0b0963b25a41cc40061269e5",
             input={"temperature": 0.75, "max_length": 500, "top_p": 1},

embedchain/apps/OpenSourceApp.py

@@ -55,6 +55,9 @@ class OpenSourceApp(EmbedChain):
                 "OpenSourceApp does not support switching models at runtime. Please create a new app instance."
             )
 
+        if config.system_prompt:
+            raise ValueError("OpenSourceApp does not support `system_prompt`")
+
         response = self.instance.generate(
             prompt=prompt,
             streaming=config.stream,