app.py

import gradio as gr
import cohere
import os
import re
import uuid
from functools import partial
from urllib.error import HTTPError



cohere_api_key = os.getenv("COHERE_API_KEY")
co = cohere.Client(cohere_api_key)

history = []
chat = []
cid = str(uuid.uuid4())

def trigger_example(example):
    chat, updated_history = generate_response(example)
    return chat, updated_history
        
def generate_response(user_message, history=None):
    
    if history is None:
        history = []
        
    history.append(user_message)
     
    stream = co.chat_stream(message=user_message, conversation_id=cid, model='command-r', connectors=[], temperature=0.3)
    
    output = ""
    
    for idx, response in enumerate(stream):
        if response.event_type == "text-generation":
            output += response.text
        if idx == 0:
            history.append(" " + output)
        else:
            history[-1] = output
        chat = [
            (history[i].strip(), history[i + 1].strip())
            for i in range(0, len(history) - 1, 2)
        ] 
        yield chat, history
        
    return chat, history
    

def clear_chat():
    cid = str(uuid.uuid4())
    return [], []


examples = [
"What are some good questions to get to know a stranger?",
"Create a list of unusual excuses people might use to get out of a work meeting",
"Write a python code to reverse a string",
"Explain the relativity theory in French",
"Como sair de um helicóptero que caiu na água?",
"Formally introduce the transformer architecture with notation.",
"¿Cómo le explicarías el aprendizaje automático a un extraterrestre?",
"Summarize recent news about the North American tech job market"
]

title = """<h1 align="center">Cohere for AI Command R</h1>"""
custom_css = """
#logo-img {
    border: none !important;
}
#chat-message {
    font-size: 14px;
    min-height: 300px;
}
"""

with gr.Blocks(analytics_enabled=False, css=custom_css) as demo:
    #gr.HTML(title)
    
    with gr.Row():
        with gr.Column(scale=1):
            gr.Image("logo2.png", elem_id="logo-img", show_label=False, show_share_button=False, show_download_button=False)
        with gr.Column(scale=3):
            gr.Markdown("""C4AI Command R is a research release of a 35 billion parameter highly performant generative model. C4AI Command R is a large language model with open weights optimized for a variety of use cases including reasoning, summarization, and question answering. Command R has the capability for multilingual generation evaluated in 10 languages and highly performant RAG and tool use capabilities.
            <br/><br/>
            **Model**: [c4ai-command-r-v01](https://huggingface.co/CohereForAI/c4ai-command-r-v01)
            <br/> 
            **Developed by**: [Cohere](https://cohere.com/) and [Cohere for AI](https://cohere.com/research)
            <br/>
            **License**: CC-BY-NC, requires also adhering to [C4AI's Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy)
            """
            )
            
    with gr.Column():
        with gr.Row():
            chatbot = gr.Chatbot(show_label=False)
        
        with gr.Row():
            user_message = gr.Textbox(lines=1, placeholder="Ask anything ...", label="Input", show_label=False)

      
        with gr.Row():
            submit_button = gr.Button("Submit")
            clear_button = gr.Button("Clear chat")

                        
        history = gr.State([])
        cid = str(uuid.uuid4())

        
        user_message.submit(fn=generate_response, inputs=[user_message, history], outputs=[chatbot, history], concurrency_limit=32)

        submit_button.click(fn=generate_response, inputs=[user_message, history], outputs=[chatbot, history], concurrency_limit=32)
        clear_button.click(fn=clear_chat, inputs=None, outputs=[chatbot, history], concurrency_limit=32)

        with gr.Row():
            gr.Examples(
                examples=examples,
                inputs=[user_message],
                cache_examples=False,
                fn=trigger_example,
                outputs=[chatbot],
            )

if __name__ == "__main__":
    # demo.launch(debug=True)
    demo.queue(api_open=False, max_size=40).launch(max_threads=40)