use existing ChatOpenAI instead of Unbound class and remove loadDotEnv

src/utils/llm.py

@@ -1,7 +1,5 @@
 from openai import OpenAI
 import pdb
-import os
-from dotenv import load_dotenv
 from langchain_openai import ChatOpenAI
 from langchain_core.globals import get_llm_cache
 from langchain_core.language_models.base import (
@@ -31,9 +29,6 @@ from langchain_ollama import ChatOllama
 from langchain_core.output_parsers.base import OutputParserLike
 from langchain_core.runnables import Runnable, RunnableConfig
 from langchain_core.tools import BaseTool
-from pydantic import Field, PrivateAttr
-import requests
-import urllib3
 
 from typing import (
     TYPE_CHECKING,
@@ -108,72 +103,6 @@ class DeepSeekR1ChatOpenAI(ChatOpenAI):
         return AIMessage(content=content, reasoning_content=reasoning_content)
 
 
-# Load environment variables
-load_dotenv()
-
-class UnboundChatOpenAI(ChatOpenAI):
-    """Chat model that uses Unbound's API."""
-    
-    _session: requests.Session = PrivateAttr()
-    
-    def __init__(self, *args: Any, **kwargs: Any) -> None:
-        kwargs["base_url"] = kwargs.get("base_url", os.getenv("UNBOUND_ENDPOINT", "https://api.getunbound.ai"))
-        kwargs["api_key"] = kwargs.get("api_key", os.getenv("UNBOUND_API_KEY"))
-        if not kwargs["api_key"]:
-            raise ValueError("UNBOUND_API_KEY environment variable is not set")
-        super().__init__(*args, **kwargs)
-        
-        self.client = OpenAI(
-            base_url=kwargs["base_url"],
-            api_key=kwargs["api_key"]
-        )
-        
-        urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
-        
-        self._session = requests.Session()
-        self._session.verify = False
-        
-    def invoke(
-        self,
-        input: LanguageModelInput,
-        config: Optional[RunnableConfig] = None,
-        *,
-        stop: Optional[list[str]] = None,
-        **kwargs: Any,
-    ) -> AIMessage:
-        message_history = []
-        for input_ in input:
-            if isinstance(input_, SystemMessage):
-                message_history.append({"role": "system", "content": input_.content})
-            elif isinstance(input_, AIMessage):
-                message_history.append({"role": "assistant", "content": input_.content})
-            else:
-                message_history.append({"role": "user", "content": input_.content})
-        
-        response = self._session.post(
-            f"{self.client.base_url}/v1/chat/completions",
-            headers={"Authorization": f"Bearer {self.client.api_key}", "Content-Type": "application/json"},
-            json={
-                "model": self.model_name or "gpt-4o-mini",
-                "messages": message_history
-            }
-        )
-        response.raise_for_status()
-        data = response.json()
-        content = data["choices"][0]["message"]["content"]
-        return AIMessage(content=content)
-    
-    async def ainvoke(
-        self,
-        input: LanguageModelInput,
-        config: Optional[RunnableConfig] = None,
-        *,
-        stop: Optional[list[str]] = None,
-        **kwargs: Any,
-    ) -> AIMessage:
-        return self.invoke(input, config, stop=stop, **kwargs)
-
-
 class DeepSeekR1ChatOllama(ChatOllama):
 
     async def ainvoke(

src/utils/utils.py

@@ -14,7 +14,7 @@ from langchain_google_genai import ChatGoogleGenerativeAI
 from langchain_ollama import ChatOllama
 from langchain_openai import AzureChatOpenAI, ChatOpenAI
 
-from .llm import DeepSeekR1ChatOpenAI, DeepSeekR1ChatOllama, UnboundChatOpenAI
+from .llm import DeepSeekR1ChatOpenAI, DeepSeekR1ChatOllama
 
 PROVIDER_DISPLAY_NAMES = {
     "openai": "OpenAI",
@@ -162,15 +162,10 @@ def get_llm_model(provider: str, **kwargs):
             api_key=os.getenv("MOONSHOT_API_KEY"),
         )
     elif provider == "unbound":
-        if not kwargs.get("base_url", ""):
+        return ChatOpenAI(
-            base_url = os.getenv("UNBOUND_ENDPOINT", "https://api.getunbound.ai")
-        else:
-            base_url = kwargs.get("base_url")
-
-        return UnboundChatOpenAI(
             model=kwargs.get("model_name", "gpt-4o-mini"),
             temperature=kwargs.get("temperature", 0.0),
-            base_url=base_url,
+            base_url = os.getenv("UNBOUND_ENDPOINT", "https://api.getunbound.ai"),
             api_key=api_key,
         )
     else: