Create app.py

app.py

@@ -0,0 +1,17 @@
+from fastapi import FastAPI, Request
+from llama_cpp import Llama
+
+app = FastAPI()
+# Load your 3.6GB model
+llm = Llama(model_path="./mpt-7b-chat.ggmlv0.q4_0.bin")
+
+@app.post("/v1/chat")
+async def chat(request: Request):
+    data = await request.json()
+    # Basic API logic:
+    response = llm(data["prompt"], max_tokens=100)
+    return {"choices": [{"text": response["choices"][0]["text"]}]}
+
+@app.get("/")
+def home():
+    return {"status": "Your AI is Always Up!"}