File size: 5,075 Bytes
c551206 e5b736f c15a43b c551206 e5b736f c551206 e5b736f 2c5ba3b e5b736f c15a43b e5b736f c15a43b e5b736f 2c5ba3b e5b736f c551206 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 |
import gradio as gr
import requests
import json
import os
API_KEY = os.getenv('API_KEY')
INVOKE_URL = "https://api.nvcf.nvidia.com/v2/nvcf/pexec/functions/0e349b44-440a-44e1-93e9-abe8dcb27158"
FETCH_URL_FORMAT = "https://api.nvcf.nvidia.com/v2/nvcf/pexec/status/"
headers = {
"Authorization": f"Bearer {API_KEY}",
"Accept": "application/json",
"Content-Type": "application/json",
}
BASE_SYSTEM_MESSAGE = "I carefully provide accurate, factual, thoughtful, nuanced answers and am brilliant at reasoning."
def clear_chat(chat_history_state, chat_message):
print("Clearing chat...")
chat_history_state = []
chat_message = ''
return chat_history_state, chat_message
def user(message, history, system_message=None):
print(f"User message: {message}")
history = history or []
if system_message: # Check if a system message is provided and should be added
history.append({"role": "system", "content": system_message})
history.append({"role": "user", "content": message})
return history
def call_nvidia_api(history, max_tokens, temperature, top_p):
payload = {
"messages": history,
"temperature": temperature,
"top_p": top_p,
"max_tokens": max_tokens,
"stream": False
}
print(f"Payload enviado: {payload}") # Imprime o payload enviado
session = requests.Session()
response = session.post(INVOKE_URL, headers=headers, json=payload)
while response.status_code == 202:
request_id = response.headers.get("NVCF-REQID")
fetch_url = FETCH_URL_FORMAT + request_id
response = session.get(fetch_url, headers=headers)
response.raise_for_status()
response_body = response.json()
print(f"Payload recebido: {response_body}") # Imprime o payload recebido
if response_body["choices"]:
assistant_message = response_body["choices"][0]["message"]["content"]
history.append({"role": "assistant", "content": assistant_message})
return history
def chat(history, system_message, max_tokens, temperature, top_p, top_k, repetition_penalty):
print("Starting chat...")
updated_history = call_nvidia_api(history, max_tokens, temperature, top_p)
return updated_history, ""
def update_chatbot(message, chat_history, system_message, max_tokens, temperature, top_p):
print("Updating chatbot...")
if not chat_history or (chat_history and chat_history[-1]["role"] != "user"):
chat_history = user(message, chat_history, system_message if not chat_history else None)
else:
chat_history = user(message, chat_history)
chat_history, _ = call_nvidia_api(chat_history, max_tokens, temperature, top_p)
return chat_history
# Gradio interface setup
with gr.Blocks() as demo:
with gr.Row():
with gr.Column():
gr.Markdown("LLAMA 2 70B Free Demo")
description="""
<div style="text-align: center; font-size: 1.5em; margin-bottom: 20px;">
<strong>Explore the Capabilities of LLAMA 2 70B</strong>
</div>
<p>Llama 2 is a large language AI model capable of generating text and code in response to prompts.
</p>
<p> <strong>How to Use:</strong></p>
<ol>
<li>Enter your <strong>message</strong> in the textbox to start a conversation or ask a question.</li>
<li>Adjust the parameters in the "Additional Inputs" accordion to control the model's behavior.</li>
<li>Use the buttons below the chatbot to submit your query, clear the chat history, or perform other actions.</li>
</ol>
<p> <strong>Powered by NVIDIA's cutting-edge AI API, LLAMA 2 70B offers an unparalleled opportunity to interact with an AI model of exceptional conversational ability, accessible to everyone at no cost.</strong></p>
<p> <strong>HF Created by:</strong> @artificialguybr (<a href="https://twitter.com/artificialguybr">Twitter</a>)</p>
<p> <strong>Discover more:</strong> <a href="https://artificialguy.com">artificialguy.com</a></p>
"""
gr.Markdown(description)
chatbox = gr.Textbox(label="What do you want to chat about?", placeholder="Ask me anything.", lines=3)
system_msg = gr.Textbox(BASE_SYSTEM_MESSAGE, label="System Message", placeholder="System prompt.", lines=5)
max_tokens = gr.Slider(20, 1024, label="Max Tokens", step=20, value=1024, interactive=True)
temperature = gr.Slider(0.0, 1.0, label="Temperature", step=0.1, value=0.2, interactive=True)
top_p = gr.Slider(0.0, 1.0, label="Top P", step=0.05, value=0.7, interactive=True)
chat_history_state = gr.State([])
additional_inputs = [system_msg, max_tokens, temperature, top_p]
chatbot = gr.ChatInterface(
fn=update_chatbot,
inputs=[chatbox, chat_history_state],
outputs=chat_history_state,
additional_inputs=additional_inputs,
title="LLAMA 2 70B Chatbot",
submit_btn="Submit",
clear_btn="🗑️ Clear",
textbox=chatbox,
)
chatbot.clear(
fn=clear_chat,
inputs=[chat_history_state, chatbox],
outputs=[chat_history_state, chatbox]
)
demo.launch() |