app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM, TextIteratorStreamer
import torch
from threading import Thread

model_id = "TheDrummer/Tiger-Gemma-9B-v3"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto"
)

def respond(message, history):
    # Build conversation (NO system prompt)
    messages = []
    for user_msg, bot_msg in history:
        messages.append({"role": "user", "content": user_msg})
        messages.append({"role": "assistant", "content": bot_msg})
    messages.append({"role": "user", "content": message})

    input_ids = tokenizer.apply_chat_template(
        messages,
        return_tensors="pt",
        add_generation_prompt=True
    ).to(model.device)

    streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
    
    thread = Thread(target=model.generate, kwargs=dict(
        input_ids=input_ids,
        max_new_tokens=512,
        temperature=0.7,
        do_sample=True,
        streamer=streamer
    ))
    thread.start()

    partial = ""
    for token in streamer:
        partial += token
        yield partial

gr.ChatInterface(
    fn=respond,
    title="Tiger-Gemma 9B Chat",
    description="Powered by TheDrummer/Tiger-Gemma-9B-v3",
).launch(share=True)