infer.py freezes when wan_video_dit is initialised

[Nikoviking]

I run the example command for infer.py: python infer.py --prompt "A man sits comfortably at a desk, facing the camera as if talking to a friend or family member on the screen. His gaze is focused and gentle, with a natural smile. The background is his carefully decorated personal space, with photos and a world map on the wall, conveying a sense of intimate and modern communication." --ip_image "test/input/lecun.jpg" --output "test/output/lecun.mp4"

I get back the following readout - and it stays that way:

(Stand-In) nick@DESKTOP-CRBCRG8:/mnt/c/Users/Nicho/Downloads/Stand-In$ python infer.py --prompt "A man sits comfortably at a desk, facing the camera as if talking to a friend or family member on the screen. His gaze is focused and gentle, with a natural smile. The background is his carefully decorated personal space, with photos and a world map on the wall, conveying a sense of intimate and modern communication." --ip_image "test/input/lecun.jpg" --output "test/output/lecun.mp4" Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}, 'CUDAExecutionProvider': {'sdpa_kernel': '0', 'use_tf32': '1', 'fuse_conv_bias': '0', 'prefer_nhwc': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_enable': '0', 'use_ep_level_unified_stream': '0', 'device_id': '0', 'has_user_compute_stream': '0', 'gpu_external_empty_cache': '0', 'cudnn_conv_algo_search': 'EXHAUSTIVE', 'cudnn_conv1d_pad_to_nc1d': '0', 'gpu_mem_limit': '18446744073709551615', 'gpu_external_alloc': '0', 'gpu_external_free': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'do_copy_in_default_stream': '1', 'enable_cuda_graph': '0', 'user_compute_stream': '0', 'cudnn_conv_use_max_workspace': '1'}} find model: checkpoints/antelopev2/models/antelopev2/1k3d68.onnx landmark_3d_68 ['None', 3, 192, 192] 0.0 1.0 Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}, 'CUDAExecutionProvider': {'sdpa_kernel': '0', 'use_tf32': '1', 'fuse_conv_bias': '0', 'prefer_nhwc': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_enable': '0', 'use_ep_level_unified_stream': '0', 'device_id': '0', 'has_user_compute_stream': '0', 'gpu_external_empty_cache': '0', 'cudnn_conv_algo_search': 'EXHAUSTIVE', 'cudnn_conv1d_pad_to_nc1d': '0', 'gpu_mem_limit': '18446744073709551615', 'gpu_external_alloc': '0', 'gpu_external_free': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'do_copy_in_default_stream': '1', 'enable_cuda_graph': '0', 'user_compute_stream': '0', 'cudnn_conv_use_max_workspace': '1'}} find model: checkpoints/antelopev2/models/antelopev2/2d106det.onnx landmark_2d_106 ['None', 3, 192, 192] 0.0 1.0 Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}, 'CUDAExecutionProvider': {'sdpa_kernel': '0', 'use_tf32': '1', 'fuse_conv_bias': '0', 'prefer_nhwc': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_enable': '0', 'use_ep_level_unified_stream': '0', 'device_id': '0', 'has_user_compute_stream': '0', 'gpu_external_empty_cache': '0', 'cudnn_conv_algo_search': 'EXHAUSTIVE', 'cudnn_conv1d_pad_to_nc1d': '0', 'gpu_mem_limit': '18446744073709551615', 'gpu_external_alloc': '0', 'gpu_external_free': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'do_copy_in_default_stream': '1', 'enable_cuda_graph': '0', 'user_compute_stream': '0', 'cudnn_conv_use_max_workspace': '1'}} find model: checkpoints/antelopev2/models/antelopev2/genderage.onnx genderage ['None', 3, 96, 96] 0.0 1.0 Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}, 'CUDAExecutionProvider': {'sdpa_kernel': '0', 'use_tf32': '1', 'fuse_conv_bias': '0', 'prefer_nhwc': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_enable': '0', 'use_ep_level_unified_stream': '0', 'device_id': '0', 'has_user_compute_stream': '0', 'gpu_external_empty_cache': '0', 'cudnn_conv_algo_search': 'EXHAUSTIVE', 'cudnn_conv1d_pad_to_nc1d': '0', 'gpu_mem_limit': '18446744073709551615', 'gpu_external_alloc': '0', 'gpu_external_free': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'do_copy_in_default_stream': '1', 'enable_cuda_graph': '0', 'user_compute_stream': '0', 'cudnn_conv_use_max_workspace': '1'}} find model: checkpoints/antelopev2/models/antelopev2/glintr100.onnx recognition ['None', 3, 112, 112] 127.5 127.5 Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}, 'CUDAExecutionProvider': {'sdpa_kernel': '0', 'use_tf32': '1', 'fuse_conv_bias': '0', 'prefer_nhwc': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_enable': '0', 'use_ep_level_unified_stream': '0', 'device_id': '0', 'has_user_compute_stream': '0', 'gpu_external_empty_cache': '0', 'cudnn_conv_algo_search': 'EXHAUSTIVE', 'cudnn_conv1d_pad_to_nc1d': '0', 'gpu_mem_limit': '18446744073709551615', 'gpu_external_alloc': '0', 'gpu_external_free': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'do_copy_in_default_stream': '1', 'enable_cuda_graph': '0', 'user_compute_stream': '0', 'cudnn_conv_use_max_workspace': '1'}} find model: checkpoints/antelopev2/models/antelopev2/scrfd_10g_bnkps.onnx detection [1, 3, '?', '?'] 127.5 128.0 set det-size: (640, 640) FaceProcessor initialized successfully. Loading models from: ['checkpoints/base_model/diffusion_pytorch_model-00001-of-00006.safetensors', 'checkpoints/base_model/diffusion_pytorch_model-00002-of-00006.safetensors', 'checkpoints/base_model/diffusion_pytorch_model-00003-of-00006.safetensors', 'checkpoints/base_model/diffusion_pytorch_model-00004-of-00006.safetensors', 'checkpoints/base_model/diffusion_pytorch_model-00005-of-00006.safetensors', 'checkpoints/base_model/diffusion_pytorch_model-00006-of-00006.safetensors'] model_name: wan_video_dit model_class: WanModel This model is initialized with extra kwargs: {'has_image_input': False, 'patch_size': [1, 2, 2], 'in_dim': 16, 'dim': 5120, 'ffn_dim': 13824, 'freq_dim': 256, 'text_dim': 4096, 'out_dim': 16, 'num_heads': 40, 'num_layers': 40, 'eps': 1e-06}

There is no output or spike in GPU activity. (RTX 4090 (24G VRAM) with 64GB RAM).
How could I fix this?

[KBRASK]

It seems that you're stuck at the stage of loading the model into GPU memory. Since Wan2.1 14B itself may exceed 24GB of memory, it cannot run due to insufficient GPU memory. The community member kijai has provided a quantization flow that allows Stand-In to run with low GPU memory.