app.py

"""
# Inference

import gradio as gr

app = gr.load(
    "google/gemma-2-2b-it",
    src = "models",
    inputs = [gr.Textbox(label = "Input")],
    outputs = [gr.Textbox(label = "Output")],
    title = "Google Gemma",
    description = "Inference",
    examples = [
        ["Hello, World."]
    ]
).launch()
"""
"""
# Pipeline

import gradio as gr
from transformers import pipeline

pipe = pipeline(model = "google/gemma-2-2b-it")

def fn(input):
    output = pipe(
        input,
        max_new_tokens = 2048
    )
    return output[0]["generated_text"]#[len(input):]

app = gr.Interface(
    fn = fn,
    inputs = [gr.Textbox(label = "Input")],
    outputs = [gr.Textbox(label = "Output")],
    title = "Google Gemma",
    description = "Pipeline",
    examples = [
        ["Hello, World."]
    ]
).launch()
"""

import gradio as gr
from huggingface_hub import InferenceClient
import os

#hf_token = os.getenv("HF_TOKEN")

client = InferenceClient("google/gemma-2-2b-it")

def respond(
    message,
    history: list[tuple[str, str]],
    #system_message,
    ##user_message,
    max_tokens,
    temperature,
    top_p,
):
    #messages = [{"role": "system", "content": system_message}]
    ##messages = [{"role": "user", "content": user_message}]
    messages = []

    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    messages.append({"role": "user", "content": message})

    response = ""

    for message in client.chat_completion(
        messages,
        max_tokens=max_tokens,
        temperature=temperature,
        top_p=top_p,
        stream=True,
    ):
        token = message.choices[0].delta.content

        response += token
        yield response

demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        #gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
    ],
)

if __name__ == "__main__":
    demo.launch()

"""
client = InferenceClient(api_key=hf_token)

def fn(prompt, history=[]):
    messages = []
    
    for user_prompt, bot_response in history:
        messages.append({"role": "user", "content": user_prompt})
        messages.append({"role": "bot", "content": bot_response})
    
    messages.append({"role": "user", "content": prompt})

    stream = client.chat.completions.create(
        model = "google/gemma-2-2b-it", 
        messages = messages, 
        #temperature = 0.5,
        #max_tokens = 2048,
        #top_p = 0.7,
        stream = True
    )

    bot_response = "".join(chunk.choices[0].delta.content for chunk in stream)

    history.append((prompt, bot_response))
    return bot_response, history

app = gr.Interface(
    fn = fn, 
    inputs = [gr.Textbox(label = "Input")],
    outputs = [gr.Textbox(label = "Output")],
    title = "Google Gemma",
    description = "Chatbot",
    examples = [
        ["Hello, World."]
    ]
).launch()
"""