import os
import gradio as gr
import spaces
from huggingface_hub import InferenceClient

MODEL = "huihui-ai/Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated"

client = InferenceClient(
    api_key=os.environ["HF_TOKEN"]
)

@spaces.GPU
def chat(message, history):
    messages = []

    history = history or []

    for user, assistant in history:
        messages.append({"role": "user", "content": user})
        messages.append({"role": "assistant", "content": assistant})

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

    response = ""

    stream = client.chat.completions.create(
        model=MODEL,
        messages=messages,
        stream=True,
        max_tokens=2048,
    )

    for chunk in stream:
        delta = chunk.choices[0].delta.content
        if delta:
            response += delta
            yield response

demo = gr.ChatInterface(
    fn=chat,
    title="Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated",
    description="Powered by Hugging Face Inference Providers",
)

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