Upload 2 files

app.py

@@ -2,6 +2,11 @@ import gradio as gr
 import spaces
 from transformers import AutoTokenizer, AutoModelForCausalLM
 import torch
+import os
+
+# Set cache to persistent storage
+os.environ['TRANSFORMERS_CACHE'] = '/data/huggingface'
+os.environ['HF_HOME'] = '/data/huggingface'
 
 # Load model and tokenizer
 model_id = "Equall/SaulLM-54B-Instruct"

requirements.txt

@@ -1,7 +1,8 @@
+pydantic==2.10.6
 transformers
 torch
 spaces
 accelerate
 sentencepiece
 protobuf
-gradio==4.44.0
+gradio