Updated requirements

main.py

@@ -14,8 +14,8 @@ from pydantic import BaseModel
 from transformers import AutoModelForCausalLM, AutoTokenizer
 import uvicorn
 import torch
-import sys
 # torch.mps.empty_cache()
+# torch.set_num_threads(1)
 
 app = FastAPI()
 
@@ -77,7 +77,6 @@ model, tokenizer, device = setup()
 
 @app.post("/generate/")
 async def generate_text(item: Item):
-    print(item, file=sys.stderr)
     return {"response": generate(item, model, tokenizer, device)}
 
 @app.get("/")

requirements.txt

@@ -1,5 +1,5 @@
 fastapi
-uvicorn[standard]
+uvicorn
 huggingface_hub
 pydantic
 transformers