Spaces:
Running
Running
File size: 1,991 Bytes
224b190 53c5f9b 224b190 53c5f9b 224b190 |
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 |
import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
# Load the model and tokenizer
model_name = "Smilyai-labs/Sam-reason-S2" # or your local path
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
model.eval()
# Chat function
def chat_with_model(user_input, history):
if history is None:
history = []
# Build conversation string for context
conversation = ""
for msg in history:
conversation += f"User: {msg['content']}\nSam:"
if msg['role'] == "assistant":
conversation += f" {msg['content']}\n"
conversation += f"User: {user_input}\nSam:"
# Encode and generate
inputs = tokenizer(conversation, return_tensors="pt", truncation=True, max_length=1024)
with torch.no_grad():
outputs = model.generate(
inputs.input_ids,
max_new_tokens=150,
do_sample=True,
top_k=50,
top_p=0.95,
temperature=0.7,
pad_token_id=tokenizer.eos_token_id
)
decoded = tokenizer.decode(outputs[0], skip_special_tokens=True)
response_text = decoded.split("Sam:")[-1].strip()
# Add new message to chat history
history.append({"role": "user", "content": user_input})
history.append({"role": "assistant", "content": response_text})
return "", history
# Gradio UI
def create_chatbot_interface():
with gr.Blocks() as demo:
gr.Markdown("# 💬 Chat with **Sam** (SmilyAI's Reasoning LLM 2nd generation)")
chatbot = gr.Chatbot(label="Chat", type="messages")
user_input = gr.Textbox(placeholder="Type your message...", show_label=False)
send_btn = gr.Button("Send")
send_btn.click(
chat_with_model,
inputs=[user_input, chatbot],
outputs=[user_input, chatbot]
)
return demo
# Launch
demo = create_chatbot_interface()
demo.launch()
|