Update app.py

app.py

@@ -1,24 +1,30 @@
 from fastapi import FastAPI, UploadFile, File
 from pydantic import BaseModel
 from diffusers import StableDiffusionPipeline
-import torch
+from huggingface_hub import hf_hub_download
+from fastapi.responses import StreamingResponse
 from PIL import Image
+import torch
 import io
-from fastapi.responses import StreamingResponse
 
 app = FastAPI()
 
-MODEL_PATH = "Interior.safetensors"
-LORA_PATH = "Interior_lora.safetensors"
-
-print("Loading model...")
-
-# ========================
-# ⚡ SAFE CPU CONFIG
-# ========================
-pipe = StableDiffusionPipeline.from_single_file(
-    MODEL_PATH,
-    torch_dtype=torch.float32,   # ✔ مهم جدًا
+# =========================
+# تحميل LoRA من HF
+# =========================
+LORA_PATH = hf_hub_download(
+    repo_id="ebraam1/interior-sd-models",
+    filename="Interior_lora.safetensors"
+)
+
+print("Loading base model...")
+
+# =========================
+# ✔ FIX: استخدم pretrained model بدل single_file
+# =========================
+pipe = StableDiffusionPipeline.from_pretrained(
+    "runwayml/stable-diffusion-v1-5",
+    torch_dtype=torch.float32,
     safety_checker=None
 ).to("cpu")
 
@@ -27,16 +33,16 @@ print("Loading LoRA...")
 pipe.load_lora_weights(LORA_PATH)
 pipe.fuse_lora(lora_scale=0.8)
 
-# ⚡ speed boost
+# =========================
+# ⚡ Speed optimizations
+# =========================
 pipe.enable_attention_slicing()
 pipe.enable_vae_slicing()
 
 print("Model ready 🔥")
 
 
-# ========================
-# REQUEST MODEL
-# ========================
+# =========================
 class Prompt(BaseModel):
     prompt: str
 
@@ -48,15 +54,15 @@ def to_bytes(img):
     return buf
 
 
-# ========================
+# =========================
 # TXT2IMG
-# ========================
+# =========================
 @app.post("/txt2img")
 def generate(data: Prompt):
 
     image = pipe(
         data.prompt,
-        num_inference_steps=6,   # ⚡ سريع
+        num_inference_steps=6,
         guidance_scale=5,
         height=256,
         width=256
@@ -65,14 +71,14 @@ def generate(data: Prompt):
     return StreamingResponse(to_bytes(image), media_type="image/png")
 
 
-# ========================
-# IMG2IMG
-# ========================
+# =========================
+# IMG2IMG (correct way)
+# =========================
 @app.post("/img2img")
 async def img2img_api(file: UploadFile = File(...), prompt: str = ""):
 
     img = Image.open(io.BytesIO(await file.read())).convert("RGB")
-    img = img.resize((256, 256))  # ⚡ أسرع
+    img = img.resize((256, 256))
 
     image = pipe(
         prompt=prompt,