Spaces:
Sleeping
Sleeping
File size: 1,625 Bytes
87e8c3e 8642e97 87e8c3e 8642e97 c9180f1 87e8c3e 8642e97 87e8c3e 8642e97 87e8c3e 8642e97 87e8c3e 8642e97 87e8c3e 8642e97 87e8c3e 5780d63 87e8c3e 8642e97 87e8c3e |
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 |
import gradio as gr
from huggingface_hub import InferenceClient
# ←–– set this to the exact name of your HF repo
HF_MODEL_ID = "rieon/DeepCoder-14B-Preview-Suger"
# explicitly tell the client you want text-generation
client = InferenceClient(model=HF_MODEL_ID)
def respond(
message: str,
history: list[tuple[str, str]],
system_message: str,
max_tokens: int,
temperature: float,
top_p: float,
):
# assemble a single prompt from system message + history
prompt = system_message.strip() + "\n"
for user, bot in history:
prompt += f"User: {user}\nAssistant: {bot}\n"
prompt += f"User: {message}\nAssistant:"
# stream back tokens
generated = ""
for chunk in client.text_generation(
inputs=prompt,
max_new_tokens=max_tokens,
temperature=temperature,
top_p=top_p,
stream=True,
):
# the API returns a small JSON with .generated_text
generated += chunk.generated_text
yield generated
demo = gr.ChatInterface(
respond,
title="DeepCoder with Suger",
description="Upload any text or pdf files and ask questions about them!",
textbox=gr.MultimodalTextbox(file_types=[".pdf", ".txt"]),
multimodal=True,
additional_inputs=[
gr.Textbox(value="You are a helpful coding assistant.", label="System message"),
gr.Slider(1, 2048, value=512, step=1, label="Max new tokens"),
gr.Slider(0.1, 4.0, value=0.7, step=0.1, label="Temperature"),
gr.Slider(0.1, 1.0, value=0.95, step=0.05, label="Top-p"),
],
)
if __name__ == "__main__":
demo.launch()
|