Update app.py

app.py

@@ -50,7 +50,7 @@ def install_dependencies(enable_optimization=False):
                 env={"FLASH_ATTENTION_SKIP_CUDA_BUILD": "TRUE"},
                 shell=True,
             )
-            
+
         # install apex for fused layernorm
         if not _is_package_available("apex"):
             subprocess.run(
@@ -100,7 +100,8 @@ def build_models(model_type, config, enable_optimization=False):
     # handle model download logic in HuggingFace Space
     from opensora.models.stdit.stdit3 import STDiT3
 
-    stdit = STDiT3.from_pretrained(HF_STDIT_MAP[model_type])
+    model_kwargs = {k: v for k, v in config.model.items() if k not in ("type", "from_pretrained", "force_huggingface")}
+    stdit = STDiT3.from_pretrained(HF_STDIT_MAP[model_type], **model_kwargs)
     stdit = stdit.cuda()
 
     # build scheduler
@@ -183,7 +184,6 @@ from opensora.utils.inference_utils import (
     prepare_multi_resolution_info,
     refine_prompts_by_openai,
     split_prompt,
-    has_openai_key
 )
 from opensora.utils.misc import to_torch_dtype
 
@@ -513,13 +513,15 @@ def main():
         with gr.Row():
             with gr.Column():
                 prompt_text = gr.Textbox(label="Prompt", placeholder="Describe your video here", lines=4)
-                refine_prompt = gr.Checkbox(value=has_openai_key(), label="Refine prompt with GPT4o", interactive=has_openai_key())
+                refine_prompt = gr.Checkbox(
+                    value=has_openai_key(), label="Refine prompt with GPT4o", interactive=has_openai_key()
+                )
                 random_prompt_btn = gr.Button("Random Prompt By GPT4o", interactive=has_openai_key())
 
                 gr.Markdown("## Basic Settings")
                 resolution = gr.Radio(
                     choices=["144p", "240p", "360p", "480p", "720p"],
-                    value="240p",
+                    value="480p",
                     label="Resolution",
                 )
                 aspect_ratio = gr.Radio(
@@ -646,6 +648,7 @@ def main():
         random_prompt_btn.click(fn=generate_random_prompt, outputs=prompt_text)
 
     # launch
+    demo.queue(max_size=5, default_concurrency_limit=1)
     demo.launch(server_port=args.port, server_name=args.host, share=args.share, max_threads=1)