|
import gradio as gr |
|
from llama_cpp import Llama |
|
from huggingface_hub import hf_hub_download |
|
import random |
|
|
|
|
|
model_path = hf_hub_download( |
|
repo_id="AstroMLab/AstroSage-8B-GGUF", |
|
filename="AstroSage-8B-Q8_0.gguf" |
|
) |
|
|
|
llm = Llama( |
|
model_path=model_path, |
|
n_ctx=2048, |
|
n_threads=4, |
|
chat_format="llama-3", |
|
seed=42, |
|
f16_kv=True, |
|
logits_all=False, |
|
use_mmap=True, |
|
use_gpu=True |
|
) |
|
|
|
|
|
GREETING_MESSAGES = [ |
|
"Greetings! I am AstroSage, your guide to the cosmos. What would you like to explore today?", |
|
"Welcome to our cosmic journey! I am AstroSage. How may I assist you in understanding the universe?", |
|
"AstroSage here. Ready to explore the mysteries of space and time. How may I be of assistance?", |
|
"The universe awaits! I'm AstroSage. What astronomical wonders shall we discuss?", |
|
] |
|
|
|
def get_random_greeting(): |
|
return random.choice(GREETING_MESSAGES) |
|
|
|
def respond(message, history, system_message, max_tokens, temperature, top_p): |
|
messages = [{"role": "system", "content": system_message}] |
|
for user_msg, assistant_msg in history: |
|
if user_msg: |
|
messages.append({"role": "user", "content": user_msg}) |
|
if assistant_msg: |
|
messages.append({"role": "assistant", "content": assistant_msg}) |
|
messages.append({"role": "user", "content": message}) |
|
|
|
try: |
|
response = llm.create_chat_completion( |
|
messages=messages, |
|
max_tokens=max_tokens, |
|
temperature=temperature, |
|
top_p=top_p |
|
) |
|
return response["choices"][0]["message"]["content"] |
|
except Exception as e: |
|
return f"Error: {e}" |
|
|
|
def clear_context(): |
|
return [], get_random_greeting() |
|
|
|
with gr.Blocks() as demo: |
|
gr.HTML("<div class='header-text'>🌌 AstroSage-LLAMA-3.1-8B</div><div class='subheader'>Your AI Guide to the Cosmos</div>") |
|
|
|
chatbot = gr.Chatbot(height=400) |
|
msg = gr.Textbox(placeholder="Ask about astronomy, astrophysics, or cosmology...") |
|
|
|
with gr.Accordion("Advanced Settings", open=False) as advanced_settings: |
|
system_msg = gr.Textbox( |
|
value="You are AstroSage, a highly knowledgeable AI assistant specialized in astronomy, astrophysics, and cosmology. Provide accurate, engaging, and educational responses about space science and the universe.", |
|
label="System Message", |
|
lines=3 |
|
) |
|
max_tokens = gr.Slider(1, 2048, value=512, step=1, label="Max Tokens") |
|
temperature = gr.Slider(0.1, 4.0, value=0.7, step=0.1, label="Temperature") |
|
top_p = gr.Slider(0.1, 1.0, value=0.9, step=0.05, label="Top-p") |
|
|
|
with gr.Row(): |
|
clear = gr.Button("🌟 New Chat") |
|
submit = gr.Button("Send 🚀", variant="primary") |
|
|
|
def handle_submit(message, history, system_message, max_tokens, temperature, top_p): |
|
response = respond(message, history, system_message, max_tokens, temperature, top_p) |
|
history.append((message, response)) |
|
return history, "" |
|
|
|
submit.click( |
|
handle_submit, |
|
[msg, chatbot, system_msg, max_tokens, temperature, top_p], |
|
[chatbot, msg] |
|
) |
|
|
|
clear.click(clear_context, [], [chatbot, msg]) |
|
|
|
if __name__ == "__main__": |
|
demo.launch() |
|
|
