update model

agent.py

@@ -7,22 +7,31 @@ from langchain_core.messages import HumanMessage, AIMessage, SystemMessage
 from langchain_core.tools import tool
 from ddgs import DDGS
 from dotenv import load_dotenv
+from groq import Groq
+from langchain_groq import ChatGroq
+from langchain_community.document_loaders.image import UnstructuredImageLoader
 
 
 load_dotenv()
 
 # Base Hugging Face LLM used by the chat wrapper
-base_llm = HuggingFaceEndpoint(
-    repo_id="deepseek-ai/DeepSeek-R1-0528",
-    # deepseek-ai/DeepSeek-OCR:novita
-    task="text-generation",
-    temperature=0.0,
-    huggingfacehub_api_token=os.getenv("HUGGINGFACEHUB_API_TOKEN"),
+# base_llm = HuggingFaceEndpoint(
+#     repo_id="openai/gpt-oss-20b:hyperbolic",
+#     # deepseek-ai/DeepSeek-OCR:novita
+#     task="text-generation",
+#     temperature=0.0,
+#     huggingfacehub_api_token=os.getenv("HUGGINGFACEHUB_API_TOKEN"),
+# )
+
+model = ChatGroq(
+    model="meta-llama/llama-4-scout-17b-16e-instruct",
+    temperature=0,
+    max_tokens=None,
+    timeout=None,
+    max_retries=2,
+    # other params...
 )
 
-# Chat model that works with LangGraph
-model = ChatHuggingFace(llm=base_llm)
-
 @tool
 def web_search(keywords: str, max_results:int = 5) -> str:
     """
@@ -49,21 +58,26 @@ def web_search(keywords: str, max_results:int = 5) -> str:
             output += f"Results: {result['title']}\n{result['body']}\n{result['href']}\n\n"
         return(output)
 
-@tool
-def get_image_file(task_id):
-    """
-    Get the image file from the question
-    Use cases:
-     - Extract Image from the question
+# @tool
+# def get_image_file(task_id):
+#     """
+#     Get the image file from the question
+#     Use cases:
+#      - Extract Image from the question
 
-     Args:
-        task_id: the task_id of the question
+#      Args:
+#         task_id: the task_id of the question
 
-    Returns:
-        Image file result
-    """
-    pass
-    return ''
+#     Returns:
+#         Image file result
+#     """
+
+#     loader = UnstructuredImageLoader("./example_data/layout-parser-paper-screenshot.png")
+
+#     data = loader.load()
+
+#     data[0]
+#     return ''
 
 
 class AgentState(TypedDict):

app.py

@@ -80,8 +80,8 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     results_log = []
     answers_payload = []
     # print(f"Running agent on {len(questions_data)} questions...")
-    print(f"Running agent on {len(questions_data[:5])} questions temporarily...")
-    for item in questions_data[:5]:
+    print(f"Running agent on {len(questions_data)} questions temporarily...")
+    for item in questions_data:
         task_id = item.get("task_id")
         question_text = item.get("question")
         if not task_id or question_text is None:

requirements.txt

@@ -18,4 +18,6 @@ pgvector
 python-dotenv
 pandas
 numpy
-ddgs
+ddgs
+groq
+unstructured[all-docs]