app.py

import os
from collections.abc import Iterator
from threading import Thread
import gradio as gr
import spaces
import torch
import edge_tts
import asyncio
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
from transformers import Qwen2VLForConditionalGeneration, AutoProcessor, TextIteratorStreamer
from transformers.image_utils import load_image
import time

DESCRIPTION = """
# QwQ Edge 💬
"""

css = '''
h1 {
  text-align: center;
  display: block;
}

#duplicate-button {
  margin: auto;
  color: #fff;
  background: #1565c0;
  border-radius: 100vh;
}
'''

MAX_MAX_NEW_TOKENS = 2048
DEFAULT_MAX_NEW_TOKENS = 1024
MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")

# Load the text-only model and tokenizer
model_id = "prithivMLmods/FastThink-0.5B-Tiny"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype=torch.bfloat16,
)
model.eval()

TTS_VOICES = [
    "en-US-JennyNeural",  # @tts1
    "en-US-GuyNeural",    # @tts2
    "en-US-AriaNeural",   # @tts3
    "en-US-DavisNeural",  # @tts4
    "en-US-JaneNeural",   # @tts5
    "en-US-JasonNeural",  # @tts6
]

# Load the multimodal (OCR) model and processor
MODEL_ID = "prithivMLmods/Qwen2-VL-OCR-2B-Instruct" 
processor = AutoProcessor.from_pretrained(MODEL_ID, trust_remote_code=True)
model_m = Qwen2VLForConditionalGeneration.from_pretrained(
    MODEL_ID,
    trust_remote_code=True,
    torch_dtype=torch.float16
).to("cuda").eval()

async def text_to_speech(text: str, voice: str, output_file="output.mp3"):
    """Convert text to speech using Edge TTS and save as MP3"""
    communicate = edge_tts.Communicate(text, voice)
    await communicate.save(output_file)
    return output_file

@spaces.GPU
def generate(
    input_dict: dict,
    chat_history: list[dict],
    max_new_tokens: int = 1024,
    temperature: float = 0.6,
    top_p: float = 0.9,
    top_k: int = 50,
    repetition_penalty: float = 1.2,
):
    """
    Generates chatbot response and handles TTS requests with multimodal input support.
    If the query starts with a TTS command (e.g. '@tts1'), the chat history is cleared
    to avoid non-text responses (like Audio) interfering with template rendering.
    """
    text = input_dict["text"]
    files = input_dict.get("files", [])

    # Check if input includes image(s)
    if len(files) > 1:
        images = [load_image(image) for image in files]
    elif len(files) == 1:
        images = [load_image(files[0])]
    else:
        images = []

    # Check if the message is for TTS
    tts_prefix = "@tts"
    is_tts = any(text.strip().lower().startswith(f"{tts_prefix}{i}") for i in range(1, 7))
    voice_index = next((i for i in range(1, 7) if text.strip().lower().startswith(f"{tts_prefix}{i}")), None)
    
    if is_tts and voice_index:
        voice = TTS_VOICES[voice_index - 1]
        text = text.replace(f"{tts_prefix}{voice_index}", "").strip()
        # Clear conversation history to avoid issues with non-text outputs.
        conversation = [{"role": "user", "content": text}]
    else:
        voice = None
        text = text.replace(tts_prefix, "").strip()
        conversation = [*chat_history, {"role": "user", "content": text}]

    # If there are images, process multimodal input
    if images:
        messages = [
            {"role": "user", "content": [
                *[{"type": "image", "image": image} for image in images],
                {"type": "text", "text": text},
            ]}
        ]
        prompt = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
        inputs = processor(text=[prompt], images=images, return_tensors="pt", padding=True).to("cuda")

        # Handle generation for multimodal input using model_m
        streamer = TextIteratorStreamer(processor, skip_prompt=True, skip_special_tokens=True)
        generation_kwargs = dict(inputs, streamer=streamer, max_new_tokens=max_new_tokens)
        thread = Thread(target=model_m.generate, kwargs=generation_kwargs)
        thread.start()

        buffer = ""
        yield "Thinking..."
        for new_text in streamer:
            buffer += new_text
            buffer = buffer.replace("<|im_end|>", "")
            time.sleep(0.01)
            yield buffer

    else:
        # Process text-only input using model
        input_ids = tokenizer.apply_chat_template(conversation, add_generation_prompt=True, return_tensors="pt")
        if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
            input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
            gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
        input_ids = input_ids.to(model.device)

        streamer = TextIteratorStreamer(tokenizer, timeout=20.0, skip_prompt=True, skip_special_tokens=True)
        generate_kwargs = dict(
            {"input_ids": input_ids},
            streamer=streamer,
            max_new_tokens=max_new_tokens,
            do_sample=True,
            top_p=top_p,
            top_k=top_k,
            temperature=temperature,
            num_beams=1,
            repetition_penalty=repetition_penalty,
        )
        t = Thread(target=model.generate, kwargs=generate_kwargs)
        t.start()

        outputs = []
        for new_text in streamer:
            outputs.append(new_text)
            yield "".join(outputs)

        final_response = "".join(outputs)

        # Yield text response first.
        yield final_response

        # If TTS was requested, yield audio output separately.
        if is_tts and voice:
            output_file = asyncio.run(text_to_speech(final_response, voice))
            yield gr.Audio(output_file, autoplay=True)

demo = gr.ChatInterface(
    fn=generate,
    additional_inputs=[
        gr.Slider(label="Max new tokens", minimum=1, maximum=MAX_MAX_NEW_TOKENS, step=1, value=DEFAULT_MAX_NEW_TOKENS),
        gr.Slider(label="Temperature", minimum=0.1, maximum=4.0, step=0.1, value=0.6),
        gr.Slider(label="Top-p (nucleus sampling)", minimum=0.05, maximum=1.0, step=0.05, value=0.9),
        gr.Slider(label="Top-k", minimum=1, maximum=1000, step=1, value=50),
        gr.Slider(label="Repetition penalty", minimum=1.0, maximum=2.0, step=0.05, value=1.2),
    ],
    examples=[
        ["@tts1 Who is Nikola Tesla, and why did he die?"],
        [{"text": "Extract JSON from the image", "files": ["examples/document.jpg"]}],
        [{"text": "summarize the letter", "files": ["examples/1.png"]}],
        ["A train travels 60 kilometers per hour. If it travels for 5 hours, how far will it travel in total?"],
        ["Write a Python function to check if a number is prime."],
        ["@tts2 What causes rainbows to form?"],
    ],
    cache_examples=False,
    type="messages",
    description=DESCRIPTION,
    css=css,
    fill_height=True,
    textbox=gr.MultimodalTextbox(label="Query Input", file_types=["image"], file_count="multiple"),
    stop_btn="Stop Generation",
    multimodal=True,
)

if __name__ == "__main__":
    demo.queue(max_size=20).launch(share=True)