app.py

def chat(message, history):
    messages = [{"role": "system", "content": "You are a lightning-fast assistant."}]
    for val in history:
        messages.append({"role": "user", "content": val[0]})
        messages.append({"role": "assistant", "content": val[1]})
    messages.append({"role": "user", "content": message})

    # FIX: Added .to(model.device) so the brain can read the input
    model_inputs = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to(model.device)
    
    streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
    
    generate_kwargs = dict(
        input_ids=model_inputs, # This is now the correct shape/type
        streamer=streamer,
        max_new_tokens=512,
        do_sample=True,
        temperature=0.7,
    )

    t = Thread(target=model.generate, kwargs=generate_kwargs)
    t.start()

    partial_message = ""
    for new_token in streamer:
        partial_message += new_token
        yield partial_message