Update app.py

app.py

@@ -0,0 +1,30 @@
+from fastapi import FastAPI
+from pydantic import BaseModel
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+app = FastAPI()
+
+# Ultra-tiny model (SAFE for free CPU)
+MODEL_NAME = "sshleifer/tiny-gpt2"
+
+tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+model = AutoModelForCausalLM.from_pretrained(MODEL_NAME)
+
+class Prompt(BaseModel):
+    message: str
+
+@app.get("/")
+def root():
+    return {"status": "TinyLLM API is running"}
+
+@app.post("/chat")
+def chat(prompt: Prompt):
+    inputs = tokenizer(prompt.message, return_tensors="pt")
+    outputs = model.generate(
+        **inputs,
+        max_new_tokens=50,
+        do_sample=True,
+        temperature=0.7
+    )
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return {"response": response}