app.py

import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "THUDM/cogvlm2-video-llama3-chat"
model = AutoModelForCausalLM.from_pretrained(model_name, trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained(model_name)

def generate_response(video_input):
    inputs = tokenizer(video_input, return_tensors="pt")
    outputs = model.generate(**inputs)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return response

iface = gr.Interface(fn=generate_response, inputs="text", outputs="text")
iface.launch()