Can't use with GPU but works with CPU?

[zaaccha-ry]

Face looks warped while using GPU - nvidia 4060, but then when using CPU the app works but much slower. CUDA installed, running 12.8 as seen under with command line
(venv) C:\Users\zacha\Downloads\Deep-Live-Cam-main>nvcc --version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2025 NVIDIA Corporation Built on Wed_Jan_15_19:38:46_Pacific_Standard_Time_2025 Cuda compilation tools, release 12.8, V12.8.61 Build cuda_12.8.r12.8/compiler.35404655_0

(venv) C:\Users\zacha\Downloads\Deep-Live-Cam-main>python -c "import torch; print(torch.version.cuda, torch.cuda.is_available())"
12.8 True

not sure if these command prompt outpout would help
Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CUDAExecutionProvider': {'device_id': '0', 'has_user_compute_stream': '0', 'cudnn_conv1d_pad_to_nc1d': '0', 'user_compute_stream': '0', 'gpu_external_alloc': '0', 'gpu_mem_limit': '18446744073709551615', 'enable_cuda_graph': '0', 'gpu_external_free': '0', 'gpu_external_empty_cache': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'cudnn_conv_algo_search': 'EXHAUSTIVE', 'do_copy_in_default_stream': '1', 'cudnn_conv_use_max_workspace': '1', 'tunable_op_enable': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0', 'prefer_nhwc': '0', 'use_ep_level_unified_stream': '0', 'use_tf32': '1', 'sdpa_kernel': '0', 'fuse_conv_bias': '0'}, 'CPUExecutionProvider': {}} find model: C:\Users\zacha/.insightface\models\buffalo_l\1k3d68.onnx landmark_3d_68 ['None', 3, 192, 192] 0.0 1.0 Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CUDAExecutionProvider': {'device_id': '0', 'has_user_compute_stream': '0', 'cudnn_conv1d_pad_to_nc1d': '0', 'user_compute_stream': '0', 'gpu_external_alloc': '0', 'gpu_mem_limit': '18446744073709551615', 'enable_cuda_graph': '0', 'gpu_external_free': '0', 'gpu_external_empty_cache': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'cudnn_conv_algo_search': 'EXHAUSTIVE', 'do_copy_in_default_stream': '1', 'cudnn_conv_use_max_workspace': '1', 'tunable_op_enable': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0', 'prefer_nhwc': '0', 'use_ep_level_unified_stream': '0', 'use_tf32': '1', 'sdpa_kernel': '0', 'fuse_conv_bias': '0'}, 'CPUExecutionProvider': {}} find model: C:\Users\zacha/.insightface\models\buffalo_l\2d106det.onnx landmark_2d_106 ['None', 3, 192, 192] 0.0 1.0 Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CUDAExecutionProvider': {'device_id': '0', 'has_user_compute_stream': '0', 'cudnn_conv1d_pad_to_nc1d': '0', 'user_compute_stream': '0', 'gpu_external_alloc': '0', 'gpu_mem_limit': '18446744073709551615', 'enable_cuda_graph': '0', 'gpu_external_free': '0', 'gpu_external_empty_cache': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'cudnn_conv_algo_search': 'EXHAUSTIVE', 'do_copy_in_default_stream': '1', 'cudnn_conv_use_max_workspace': '1', 'tunable_op_enable': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0', 'prefer_nhwc': '0', 'use_ep_level_unified_stream': '0', 'use_tf32': '1', 'sdpa_kernel': '0', 'fuse_conv_bias': '0'}, 'CPUExecutionProvider': {}} find model: C:\Users\zacha/.insightface\models\buffalo_l\det_10g.onnx detection [1, 3, '?', '?'] 127.5 128.0 Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CUDAExecutionProvider': {'device_id': '0', 'has_user_compute_stream': '0', 'cudnn_conv1d_pad_to_nc1d': '0', 'user_compute_stream': '0', 'gpu_external_alloc': '0', 'gpu_mem_limit': '18446744073709551615', 'enable_cuda_graph': '0', 'gpu_external_free': '0', 'gpu_external_empty_cache': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'cudnn_conv_algo_search': 'EXHAUSTIVE', 'do_copy_in_default_stream': '1', 'cudnn_conv_use_max_workspace': '1', 'tunable_op_enable': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0', 'prefer_nhwc': '0', 'use_ep_level_unified_stream': '0', 'use_tf32': '1', 'sdpa_kernel': '0', 'fuse_conv_bias': '0'}, 'CPUExecutionProvider': {}} find model: C:\Users\zacha/.insightface\models\buffalo_l\genderage.onnx genderage ['None', 3, 96, 96] 0.0 1.0 Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CUDAExecutionProvider': {'device_id': '0', 'has_user_compute_stream': '0', 'cudnn_conv1d_pad_to_nc1d': '0', 'user_compute_stream': '0', 'gpu_external_alloc': '0', 'gpu_mem_limit': '18446744073709551615', 'enable_cuda_graph': '0', 'gpu_external_free': '0', 'gpu_external_empty_cache': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'cudnn_conv_algo_search': 'EXHAUSTIVE', 'do_copy_in_default_stream': '1', 'cudnn_conv_use_max_workspace': '1', 'tunable_op_enable': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0', 'prefer_nhwc': '0', 'use_ep_level_unified_stream': '0', 'use_tf32': '1', 'sdpa_kernel': '0', 'fuse_conv_bias': '0'}, 'CPUExecutionProvider': {}} find model: C:\Users\zacha/.insightface\models\buffalo_l\w600k_r50.onnx recognition ['None', 3, 112, 112] 127.5 127.5 set det-size: (640, 640) Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CUDAExecutionProvider': {'device_id': '0', 'has_user_compute_stream': '0', 'cudnn_conv1d_pad_to_nc1d': '0', 'user_compute_stream': '0', 'gpu_external_alloc': '0', 'gpu_mem_limit': '18446744073709551615', 'enable_cuda_graph': '0', 'gpu_external_free': '0', 'gpu_external_empty_cache': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'cudnn_conv_algo_search': 'EXHAUSTIVE', 'do_copy_in_default_stream': '1', 'cudnn_conv_use_max_workspace': '1', 'tunable_op_enable': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0', 'prefer_nhwc': '0', 'use_ep_level_unified_stream': '0', 'use_tf32': '1', 'sdpa_kernel': '0', 'fuse_conv_bias': '0'}, 'CPUExecutionProvider': {}} inswapper-shape: [1, 3, 128, 128]

have 32 gb of ram so definitely not that as the issue, is it the cuda? already reinstalled multiple times, have vs 2022, vsc, vs2022 c++ installed, not sure what to do

[geedichtnohne]

i have the same problem

[geedichtnohne]

Had BlackBox face with CPU.
Had PizzaFace with GPU.

Fixed both by replacing [inswapper_128_fp16.onnx] that HuggingFace spit out... (size 270,790) with newer one I had from a different project
(size 541,263).

The one linked in the install instructions did NOT work.

This one worked ---v

https://huggingface.co/Devia/G/blob/main/inswapper_128.onnx

Rename to [ inswapper_128_fp16.onnx ] looks like that name hard coded.