app.py

import gradio as gr
import os
from huggingface_hub import InferenceClient
from langgraph.prebuilt import create_react_agent
from search_agent import tools
from langchain_huggingface import ChatHuggingFace, HuggingFaceEndpoint
from search_agent import tools
from langchain_core.messages import HumanMessage, AIMessage, SystemMessage
huggingfacehub_api_token = os.getenv('hf_api')

"""
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
"""
# client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
llm = HuggingFaceEndpoint(
    repo_id="meta-llama/Llama-3.2-1B-Instruct"  ,
    huggingfacehub_api_token=huggingfacehub_api_token,
)

chat_model = ChatHuggingFace(llm=llm, verbose =  True)
graph = create_react_agent(chat_model, tools=tools)


def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    messages = [{"role": "system", "content": system_message}]

    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    messages.append({"role": "user", "content": message})

    # response = ""

    # for message in client.chat_completion(
    #     messages,
    #     max_tokens=max_tokens,
    #     stream=True,
    #     temperature=temperature,
    #     top_p=top_p,
    # ):
    #     token = message.choices[0].delta.content

    #     response += token
    #     yield response
    def convert(msg):
        if msg["role"] in ["user", "human"]:
            return HumanMessage(content=msg["content"])
        elif msg["role"] in ["assistant", "ai"]:
            return AIMessage(content=msg["content"])
        elif msg["role"] == "system":
            return SystemMessage(content=msg["content"])
        else:
            raise ValueError(f"Unsupported role: {msg['role']}")
        
    inputs = {"messages": [convert(m) for m in messages]}
    
    # Get the response from the agent (this integrates your agent with the model)
    agent_response = graph.invoke(inputs)  # Process the inputs through your agent
    
    # Return the final message from the agent
    return agent_response['messages'][-1][1]


"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
    ],
)


if __name__ == "__main__":
    demo.launch()