Update app.py

app.py

@@ -1,3 +1,4 @@
+import os
 from fastapi import FastAPI, HTTPException
 from pydantic import BaseModel
 from sentence_transformers import SentenceTransformer
@@ -7,7 +8,10 @@ import os
 import logging
 from groq import Groq
 from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
-
+# ✅ Set a writable cache directory
+os.environ["HF_HOME"] = "/tmp/huggingface"
+os.environ["TRANSFORMERS_CACHE"] = "/tmp/huggingface"
+os.environ["SENTENCE_TRANSFORMERS_HOME"] = "/tmp/huggingface"
 # ✅ Initialize FastAPI
 app = FastAPI()
 
@@ -19,11 +23,12 @@ if not GROQ_API_KEY:
 client = Groq(api_key=GROQ_API_KEY)  # ✅ Ensure the API key is passed correctly
 
 
-# ✅ Load AI Models
-similarity_model = SentenceTransformer("sentence-transformers/all-mpnet-base-v2")
-embedding_model = SentenceTransformer("all-MiniLM-L6-v2")
-summarization_model = AutoModelForSeq2SeqLM.from_pretrained("google/long-t5-tglobal-base")
-summarization_tokenizer = AutoTokenizer.from_pretrained("google/long-t5-tglobal-base")
+# ✅ Load AI Models (Now uses /tmp/huggingface as cache)
+similarity_model = SentenceTransformer("sentence-transformers/all-mpnet-base-v2", cache_folder="/tmp/huggingface")
+embedding_model = SentenceTransformer("all-MiniLM-L6-v2", cache_folder="/tmp/huggingface")
+summarization_model = AutoModelForSeq2SeqLM.from_pretrained("google/long-t5-tglobal-base", cache_dir="/tmp/huggingface")
+summarization_tokenizer = AutoTokenizer.from_pretrained("google/long-t5-tglobal-base", cache_dir="/tmp/huggingface")
+
 
 # ✅ Load datasets
 try: