Create app.py

app.py

@@ -0,0 +1,36 @@
+from fastapi import FastAPI
+from pydantic import BaseModel
+from lmstudio import LLM  # LM Studio Python interface
+from fastapi.middleware.cors import CORSMiddleware
+
+# FastAPI app
+app = FastAPI(title="ChatGPT OS 1.0", description="Local AI Chat API", version="1.0")
+
+# Enable CORS so frontend can connect from any origin
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_methods=["*"],
+    allow_headers=["*"]
+)
+
+# Load the GPT-OSS model
+model_path = "./models/gpt-oss-20b-Q3_K_M.gguf"
+llm = LLM(model_path=model_path, context_length=16384, flash_attention=True)
+
+# Request body schema
+class ChatRequest(BaseModel):
+    message: str
+
+# Chat endpoint
+@app.post("/chat")
+async def chat(req: ChatRequest):
+    user_message = req.message
+    # Generate response from the model
+    response = llm.generate(user_message, max_new_tokens=256)
+    return {"response": response}
+
+# Optional: health check
+@app.get("/health")
+def health():
+    return {"status": "ok", "model_loaded": True}