Update app.py

app.py

@@ -8,14 +8,14 @@ from typing import List
 app = FastAPI()
 
 # ===== CONFIGURATION =====
-MAX_BATCH_SIZE = 8        # max requests processed together
+MAX_BATCH_SIZE = 4        # max requests processed together
 QUEUE_TIMEOUT = 0.1       # wait time to collect batch
 MODEL_WORKERS = 2         # your 2 vCPU
 
 # ===== LOAD MODELS =====
 models = [
     WhisperModel(
-        "small",
+        "base",
         device="cpu",
         compute_type="int8",
         cpu_threads=1