app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_name = "lamapi/next-1b"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    device_map="auto"
)
model.eval()

def chat(message, history):
    prompt = ""
    for user, bot in history or []:
        prompt += f"<start_of_turn>user\n{user}<end_of_turn>\n<start_of_turn>model\n{bot}<end_of_turn>\n"
    prompt += f"<start_of_turn>user\n{message}<end_of_turn>\n<start_of_turn>model\n"

    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)

    with torch.inference_mode():
        outputs = model.generate(
            **inputs,
            max_new_tokens=2048,
            do_sample=True,
            temperature=0.9,
            top_k=140,
            top_p=0.9
        )

    text = tokenizer.decode(outputs[0], skip_special_tokens=True)
    reply = text.split("\nmodel")[-1].strip()
    return reply  # sadece string dönüyoruz!

iface = gr.ChatInterface(
    fn=chat,
    title="Next-1B Chatbot ⚡"
)

iface.launch(share=True, server_name="0.0.0.0", server_port=7860)