gpt

app.py

@@ -125,16 +125,16 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     print(f"Using OpenAI API key: {openai_api_key[:4]}... (truncated for security)")
 
     #NMODEL
-    #'''
+    '''
     llm_client = ChatGoogleGenerativeAI(
     model="gemini-2.0-flash",           # or another Gemini model name
     google_api_key=google_api_key, # your Gemini API key
     temperature=0,
     )
-    #'''
+    '''
 
     
-    #llm_client = ChatOpenAI(model='gpt-4o',temperature=0.1,api_key=openai_api_key)
+    llm_client = ChatOpenAI(model='gpt-4o',temperature=0,api_key=openai_api_key)
 
     tavily_api_key = os.getenv("TAVILY_API_KEY")
     if not tavily_api_key:
@@ -143,7 +143,8 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     print(f"Using Tavily API key: {tavily_api_key[:4]}... (truncated for security)")
 
     travily_api_search_tool = get_travily_api_search_tool(tavily_api_key)
-    tools = [travily_api_search_tool, repl_tool, file_saver_tool,audio_transcriber_tool,wikipedia_search_tool,wikipedia_full_content_tool]
+    #tools = [travily_api_search_tool, repl_tool, file_saver_tool,audio_transcriber_tool,wikipedia_search_tool,wikipedia_full_content_tool]
+    tools = [travily_api_search_tool, repl_tool, file_saver_tool,audio_transcriber_tool]
 
     # Pull a predefined prompt from LangChain Hub
     # "hwchase17/react-chat" is a prompt template designed for ReAct-style conversational agents.
@@ -168,25 +169,12 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     IMPORTANT NOTE ON TOOL USAGE:
     - If an 'Observation' from a tool does NOT directly contain the specific answer to your question, you MUST refine your query or switch to a different, more suitable tool (e.g., 'tavily_search' for broader or more current information if 'wikipedia_search_tool' was insufficient). Do NOT get stuck repeatedly using the same tool if it's not yielding the direct answer.
     - If the input contains the exact phrase "Attachment '{{file_name}}' available at: {{attachment_url}}" (where '{{file_name}}' and '{{attachment_url}}' are placeholders for actual values), consider the file type:
-      - If the file type is binary/text (e.g., .xlsx, .docx, .mp3, .jpg, .pdf), you MUST use the 'file_saver' tool to download and save it.
+      - If the file type is binary/text (e.g., .xlsx, .docx, .mp3, .jpg, .pdf,.png), you MUST use the 'file_saver' tool to download and save it.
         For 'file_saver', the Action Input must be a JSON string like: '{{"url": "the_attachment_url", "local_filename": "the_file_name_from_attachment"}}' 
         Example: If the attachment is 'Homework.mp3' at 'https://agents-course-unit4-scoring.hf.space/files/121898981', Action Input for file_saver would be '{{"url": "https://agents-course-unit4-scoring.hf.space/files/121898981", "local_filename": "Homework.mp3"}}'
 
     IMPORTANT: When processing audio files (like .mp3) that have been saved using 'file_saver', the 'audio_transcriber_tool' MUST be used with the 'local_filename' of the saved audio file as its Action Input. Do NOT pass URLs or remote paths directly to 'audio_transcriber_tool'.
 
-    If you need to count or extract items from a Wikipedia section (like a list of albums with years), use the 'wikipedia_full_content_tool' to get the section text, then use the 'python_repl' tool to parse the text and count the relevant items.
-    Example:
-    Thought: I need to count Mercedes Sosa's studio albums from 2000 to 2009.
-    Action: wikipedia_full_content_tool
-    Action Input: "Mercedes Sosa section: Discography"
-    Observation: [Discography text]
-    Thought: I need to parse this text and count albums released between 2000 and 2009.
-    Action: python_repl
-    Action Input: [Python code that parses the text and counts albums by year]
-    Observation: [Result]
-    Thought: I have found the answer.
-    Final Answer: [number]
-
     If you have sufficient information and can provide a CONCISE response, or if no tool is needed, you MUST use this precise format:
     if you can use a LLM to answer the question, think step-by-step and then answer the question.
     Example: given a chess board image and asked to predict the next best move, if Multi-modal LLM is available, you can use it to answer the question.
@@ -231,17 +219,17 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     # Initialize gemini model with streaming enabled
     # Streaming allows tokens to be processed in real-time, reducing response latency.
     #NMODEL
-    #'''
+    '''
     summary_llm = ChatGoogleGenerativeAI(
         model="gemini-2.0-flash",           # or another Gemini model name
         google_api_key=google_api_key, # your Gemini API key
         temperature=0,
         streaming=True
     )
-    #'''
+    '''
     
     
-    #summary_llm = ChatOpenAI(model='gpt-4o', temperature=0, streaming=True,api_key=openai_api_key)
+    summary_llm = ChatOpenAI(model='gpt-4o', temperature=0, streaming=True,api_key=openai_api_key)
 
     # Create a ReAct agent
     # The agent will reason and take actions based on retrieved tools and memory.