File size: 1,526 Bytes
3cb97b6 2ef1e0a 1782685 f12ea3a 3cb97b6 f12ea3a 3cb97b6 f12ea3a 3cb97b6 f12ea3a 3cb97b6 f12ea3a 3cb97b6 0d6a629 3cb97b6 0d6a629 3cb97b6 f12ea3a 3cb97b6 |
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 |
import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
import torch
# Load model and tokenizer with trust_remote_code=True
model_id = "PowerInfer/SmallThinker-21BA3B-Instruct"
tokenizer = AutoTokenizer.from_pretrained(
model_id,
trust_remote_code=True # Required for models with custom code
)
model = AutoModelForCausalLM.from_pretrained(
model_id,
device_map="cpu", # Run on CPU
torch_dtype=torch.float32, # Use float32 on CPU
trust_remote_code=True # Allow custom code execution
)
# Create text generation pipeline
generator = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
device=-1 # CPU
)
# Define the chat function
def chat(prompt, max_new_tokens=256, temperature=0.7):
output = generator(
prompt,
max_new_tokens=max_new_tokens,
temperature=temperature,
do_sample=True,
pad_token_id=tokenizer.eos_token_id
)
return output[0]["generated_text"]
# Launch the Gradio interface
gr.Interface(
fn=chat,
inputs=[
gr.Textbox(label="Prompt", lines=4, placeholder="Ask anything..."),
gr.Slider(32, 512, value=256, step=16, label="Max New Tokens"),
gr.Slider(0.1, 1.5, value=0.7, step=0.1, label="Temperature")
],
outputs=gr.Textbox(label="Response"),
title="💬 SmallThinker-21BA3B-Instruct",
description="Run PowerInfer/SmallThinker-21BA3B-Instruct locally on CPU using Hugging Face + Gradio"
).launch()
|