app.py

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

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

model_name = "ai4bharat/Airavata"

tokenizer = AutoTokenizer.from_pretrained(model_name, padding_side="left")
tokenizer.pad_token = tokenizer.eos_token
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16).to(device)


def create_prompt_with_chat_format(messages, bos="<s>", eos="</s>", add_bos=True):
    formatted_text = ""
    for message in messages:
        if message["role"] == "system":
            formatted_text += "<|system|>\n" + message["content"] + "\n"
        elif message["role"] == "user":
            formatted_text += "<|user|>\n" + message["content"] + "\n"
        elif message["role"] == "assistant":
            formatted_text += "<|assistant|>\n" + message["content"].strip() + eos + "\n"
        else:
            raise ValueError(
                "Tulu chat template only supports 'system', 'user' and 'assistant' roles. Invalid role: {}.".format(
                    message["role"]
                )
            )
    formatted_text += "<|assistant|>\n"
    formatted_text = bos + formatted_text if add_bos else formatted_text
    return formatted_text

def inference(input_prompt, model, tokenizer):
    input_prompt = create_prompt_with_chat_format([{"role": "user", "content": input_prompt}], add_bos=False)

    encodings = tokenizer(input_prompt, padding=True, return_tensors="pt")
    encodings = encodings.to(device)

    with torch.inference_mode():  # Add missing import statement for torch.inference_mode()
        outputs = model.generate(encodings.input_ids, do_sample=False, max_new_tokens=250)

    output_text = tokenizer.decode(outputs[0], skip_special_tokens=True)

    input_prompt = tokenizer.decode(tokenizer.encode(input_prompt), skip_special_tokens=True)
    output_text = output_text[len(input_prompt):]

    return output_text


def chat_interface(message,history):
    outputs = inference(message, model, tokenizer)
    return outputs


chat_interface = gr.ChatInterface(chat_interface, title="CAMAI")
chat_interface.launch()