Scale ZeroGPU duration by asset count

README.md

@@ -70,6 +70,8 @@ The deployed Space is fail-closed: if the primary neural model cannot run, gener
 
 ZeroGPU is free for eligible personal accounts, but it is quota-limited rather than unlimited: free accounts currently receive five minutes of GPU time per day. Queueing or a quota message is therefore possible even though no inference credits or payment are required.
 
+Initial generation uses a dynamic ZeroGPU duration estimate based on the number of separate asset roles, while one-role regeneration reserves a smaller fixed window. This changes only scheduler reservation and queue priority; it does not reduce diffusion steps or image quality. The estimator avoids rejecting short jobs merely because an unnecessarily large fixed duration exceeds the visitor's remaining quota.
+
 The primary pipeline can be configured with:
 
 ```text

app.py

@@ -158,7 +158,7 @@ PRIMARY_TEXT_PIPE = None
 PRIMARY_MODEL_ERROR = None
 
 
-def gpu_task(duration: int):
+def gpu_task(duration):
     """Use a real ZeroGPU allocation on Hugging Face and remain importable in local tests."""
     if hf_spaces is not None:
         return hf_spaces.GPU(duration=duration)
@@ -2402,7 +2402,20 @@ def empty_generation_result(message: str, html_code: str = ""):
     )
 
 
-@gpu_task(duration=120)
+def estimate_generation_gpu_duration(
+    html_code: str,
+    roles: str,
+    game_type: str,
+    perspective: str,
+    theme: str,
+) -> int:
+    """Reserve realistic ZeroGPU time based on the number of independent assets."""
+    del game_type, perspective, theme
+    role_count = max(1, len(resolve_role_lines(html_code or "", roles or "")[0]))
+    return min(120, 30 + role_count * 18)
+
+
+@gpu_task(duration=estimate_generation_gpu_duration)
 def generate_images_and_game(
     html_code: str,
     roles: str,
@@ -2475,7 +2488,7 @@ def generate_images_and_game(
     )
 
 
-@gpu_task(duration=90)
+@gpu_task(duration=45)
 def regenerate_selected_asset(state: dict, selected_role: str, approved_roles: list[str]):
     if not state or not selected_role:
         return (