Update src/model_server.py

src/model_server.py

@@ -27,7 +27,7 @@ def run_stable_diffusion(prompt: str):
 
     pipe = StableDiffusionPipeline.from_pretrained(
         "runwayml/stable-diffusion-v1-5",
-        use_auth_token=os.environ["HF_TOKEN"],
+        use_auth_token=os.environ["CLIENT_TOKEN1"],
     ).to("cuda")
 
     image = pipe(prompt, num_inference_steps=10).images[0]
@@ -365,4 +365,11 @@ def optimize(pipe, compile=True):
 
 if __name__ == "__main__":
     print("Starting Modal Flux API server...")
-    # This will be handled by Modal's deployment
+    # This will be handled by Modal's deployment
+        # Already inside the Modal container → call directly
+        result = model_instance.inference(
+            request.prompt,
+            request.num_inference_steps,
+            request.width,
+            request.height
+        )