fix

app.py

@@ -3,9 +3,6 @@ import gradio as gr
 from transformers import AutoModelForCausalLM, AutoTokenizer
 import torch
 
-# Gradio client kullanımı için gerekli kod
-from gradio_client import Client
-
 MODEL_NAME = "osmankoc/llama-2-7b-zoa"
 
 # Model ve tokenizer'ı önceden yükle
@@ -38,11 +35,3 @@ demo = gr.Interface(
 demo.add_api_route("/predict", generate)
 
 demo.launch()
-
-
-
-client = Client("osmankoc/llama2-zoa-api")
-result = client.predict(
-    prompt="Hello!!",
-    api_name="/predict"
-)

requirements.txt

@@ -2,5 +2,4 @@ fastapi
 uvicorn
 torch
 transformers
-accelerate
-gradio_client
+accelerate