Wan2GP: MIOpen Error: D:/jam/TheRock/ml-libs/MIOpen/src/ocl/convolutionocl.cpp:536: Requested convoluti…

@chubbyantics69posted 8/17/2026, 8:21:44 AM·0 replies

App: Wan2GP (wan.git)
Repo: https://github.com/pinokiofactory/wan.git
Generated: 2026-08-17T08:20:32.751Z
Pinokio: 8.0.40
Platform: win32 x64
Node: v22.21.1

Summary

System

{
  "pinokio": {
    "version": "8.0.40",
    "node": "v22.21.1",
    "platform": "win32",
    "arch": "x64"
  },
  "hardware": {
    "gpu": "amd",
    "gpu_model": "amd radeon rx 9070 xt",
    "ram_gb": 64,
    "vram_gb": 16
  },
  "os": {
    "platform": "Windows",
    "distro": "Microsoft Windows 11 Pro",
    "release": "10.0.26200",
    "codename": "25H2",
    "kernel": "10.0.26200",
    "arch": "x64",
    "build": "26200",
    "servicepack": "0.0",
    "uefi": true
  },
  "system": {
    "manufacturer": "Micro-Star International Co., Ltd.",
    "model": "MS-7C95",
    "version": "1.0",
    "virtual": false
  },
  "cpu": {
    "manufacturer": "AMD",
    "brand": "Ryzen 9 5900X 12-Core Processor",
    "vendor": "AuthenticAMD",
    "family": "25",
    "model": "33",
    "stepping": "2",
    "revision": "8450",
    "speed": 3.7,
    "speedMin": 3.7,
    "speedMax": 3.7,
    "cores": 24,
    "physicalCores": 12,
    "processors": 1,
    "performanceCores": 24,
    "efficiencyCores": 0,
    "virtualization": false,
    "cache": {
      "l1d": 384,
      "l1i": 384,
      "l2": 6291456,
      "l3": 67108864
    }
  },
  "memory": {
    "total": 68641222656,
    "free": 59872382976,
    "used": 8768839680,
    "active": 8768839680,
    "available": 59872382976,
    "buffers": 0,
    "cached": 0,
    "slab": 0,
    "buffcache": 0,
    "swaptotal": 9663676416,
    "swapused": 130023424,
    "swapfree": 9533652992
  },
  "gpus": [
    {
      "model": "amd radeon rx 9070 xt"
    }
  ],
  "graphics": {
    "controllers": [
      {
        "vendor": "Advanced Micro Devices, Inc.",
        "model": "AMD Radeon RX 9070 XT",
        "bus": "PCI",
        "vram": 16304,
        "vramDynamic": true
      }
    ],
    "displays": [
      {
        "model": "SAMSUNG",
        "main": true,
        "builtin": false,
        "connection": "HDMI",
        "currentResX": 1280,
        "currentResY": 720,
        "resolutionX": 3840,
        "resolutionY": 2160,
        "pixelDepth": 32,
        "currentRefreshRate": 29
      }
    ]
  }
}

Logs

logs/api/start.js/1786949485072

Source: api / start.js
Lines: 175 total, last 175 included

[api shell.run]
Microsoft Windows [Version 10.0.26200.9168]
(c) Microsoft Corporation. All rights reserved.

D:\pinokio\api\wan.git\app>conda_hook & conda deactivate & conda deactivate & conda deactivate & conda activate base & D:\pinokio\api\wan.git\app\venv\Scripts\activate D:\pinokio\api\wan.git\app\venv && python wgp.py --multiple-images --advanced 
[GGUF][llama.cpp CUDA] kernels unavailable, using fallback
Downloading ffmpeg-9.0.1-essentials_build.zip: 100%|███████████████████████████████████████████████████████████████████| 111M/111M [00:10<00:00, 11.1MB/s]
Loaded plugin: Motion Designer (from motion_designer)
* Running on local URL:  http://127.0.0.1:42003
* To create a public link, set `share=True` in `launch()`.
[FlashVSR] Auto backend cannot use SpargeAttn (recommended, best quality especially when there is motion): FlashVSR sparse attention requirements are not satisfied. Backend: SpargeAttn (recommended, best quality especially when there is motion). Missing: Triton, SpargeAttn. Install them from docs/INSTALLATION.md and restart WanGP. Install SpargeAttn for better FlashVSR quality.
[FlashVSR] Auto backend trying Triton Sparse Attention.
pose/dw-ll_ucoco_384.onnx: 100%|███████████████████████████████████████████████████████████████████████████████████████| 134M/134M [00:12<00:00, 10.9MB/s]
pose/yolox_l.onnx: 100%|███████████████████████████████████████████████████████████████████████████████████████████████| 217M/217M [00:18<00:00, 11.7MB/s]
scribble/netG_A_latest.pth: 100%|████████████████████████████████████████████████████████████████████████████████████| 17.2M/17.2M [00:01<00:00, 12.0MB/s]
flow/raft-things.pth: 100%|██████████████████████████████████████████████████████████████████████████████████████████| 21.1M/21.1M [00:01<00:00, 12.5MB/s]
depth/depth_anything_v2_vitl.pth: 100%|██████████████████████████████████████████████████████████████████████████████| 1.34G/1.34G [02:05<00:00, 10.7MB/s]
config.json: 1.60kB [00:00, ?B/s]
feature_extractor_config.json: 1.60kB [00:00, ?B/s]
wav2vec/model.safetensors: 100%|███████████████████████████████████████████████████████████████████████████████████████| 378M/378M [00:26<00:00, 14.5MB/s]
preprocessor_config.json: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████| 159/159 [00:00<?, ?B/s]
special_tokens_map.json: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████| 85.0/85.0 [00:00<?, ?B/s]
tokenizer_config.json: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████| 163/163 [00:00<?, ?B/s]
vocab.json: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 291/291 [00:00<?, ?B/s]
config.json: 1.95kB [00:00, ?B/s]
chinese-wav2vec2-base/pytorch_model.bin: 100%|█████████████████████████████████████████████████████████████████████████| 380M/380M [00:22<00:00, 17.1MB/s]
preprocessor_config.json: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████| 160/160 [00:00<?, ?B/s]
roformer/model_bs_roformer_ep_317_sdr_12(…): 100%|█████████████████████████████████████████████████████████████████████| 639M/639M [00:22<00:00, 28.0MB/s]
(…)odel_bs_roformer_ep_317_sdr_12.9755.yaml: 2.27kB [00:00, ?B/s]
download_checks.json: 28.3kB [00:00, ?B/s]
pyannote/pyannote_model_wespeaker-voxcel(…): 100%|███████████████████████████████████████████████████████████████████| 26.6M/26.6M [00:01<00:00, 13.8MB/s]
pyannote/pytorch_model_segmentation-3.0.(…): 100%|███████████████████████████████████████████████████████████████████| 5.91M/5.91M [00:00<00:00, 7.06MB/s]
det_align/detface.pt: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 187M/187M [00:16<00:00, 11.6MB/s]
mask/sam_vit_h_4b8939_fp16.safetensors: 100%|████████████████████████████████████████████████████████████████████████| 1.28G/1.28G [01:48<00:00, 11.8MB/s]
mask/matanyone.safetensors: 100%|██████████████████████████████████████████████████████████████████████████████████████| 141M/141M [00:14<00:00, 9.60MB/s]
config.json: 2.90kB [00:00, ?B/s]
ltx-2.5-22b-distilled_diffusion_model_nv(…): 100%|███████████████████████████████████████████████████████████████████| 14.7G/14.7G [19:45<00:00, 12.4MB/s]
ltx-2.3-22b-ic-lora-hdr-scene-emb.safete(…): 100%|███████████████████████████████████████████████████████████████████| 12.6M/12.6M [00:02<00:00, 5.76MB/s]
ltx-2.5-spatial-upscaler-x2-1.0_bf16.saf(…): 100%|█████████████████████████████████████████████████████████████████████| 996M/996M [01:16<00:00, 13.0MB/s]
ltx-2.5-temporal-upscaler-x2-1.0_bf16.sa(…): 100%|█████████████████████████████████████████████████████████████████████| 262M/262M [00:21<00:00, 12.1MB/s]
ltx-2.5-22b_video_vae_bf16.safetensors: 100%|████████████████████████████████████████████████████████████████████████| 1.45G/1.45G [01:44<00:00, 13.9MB/s]
ltx-2.5-22b_audio_vae_bf16.safetensors: 100%|██████████████████████████████████████████████████████████████████████████| 107M/107M [00:09<00:00, 11.4MB/s]
ltx-2.5-22b_vocoder_bf16.safetensors: 100%|████████████████████████████████████████████████████████████████████████████| 258M/258M [00:22<00:00, 11.4MB/s]
ltx-2.5-22b_text_embedding_projection_bf(…): 100%|███████████████████████████████████████████████████████████████████| 2.31G/2.31G [02:52<00:00, 13.4MB/s]
ltx-2.5-22b_video_embeddings_connector_n(…): 100%|███████████████████████████████████████████████████████████████████| 3.23G/3.23G [03:50<00:00, 14.0MB/s]
ltx-2.5-22b_audio_embeddings_connector_n(…): 100%|█████████████████████████████████████████████████████████████████████| 807M/807M [00:57<00:00, 14.0MB/s]
ltx-2.5-22b_diffusion_video_vae_bf16.saf(…): 100%|███████████████████████████████████████████████████████████████████| 1.47G/1.47G [01:50<00:00, 13.3MB/s]
config.json: 2.50kB [00:00, 1.65MB/s]
chat_template.jinja: 18.7kB [00:00, ?B/s]
gemma4-12b-ltx-v1/tokenizer.json: 100%|██████████████████████████████████████████████████████████████████████████████| 32.2M/32.2M [00:03<00:00, 10.2MB/s]
tokenizer_config.json: 3.74kB [00:00, ?B/s]
Loading Model 'ckpts\ltx-2.5-22b-distilled_diffusion_model_nvfp4.safetensors' ...
gemma4-12b-ltx-v1/gemma4-12b-ltx-v1_int8(…): 100%|███████████████████████████████████████████████████████████████████| 12.9G/12.9G [18:14<00:00, 11.8MB/s]
Loading Text Encoder 'ckpts\gemma4-12b-ltx-v1\gemma4-12b-ltx-v1_int8_convrot.safetensors' ...
NVFP4: kernels unavailable; using fallback.
************ Memory Management for the GPU Poor (mmgp 3.7.12) by DeepBeepMeep ************
Switching to partial pinning since full requirements for pinned models is 36972.1 MB while estimated available reservable RAM is 26184.5 MB. You may increase the value of parameter 'perc_reserved_mem_max' to a value higher than 0.40 to force full pinnning.
Partial pinning of data of 'transformer' to reserved RAM
The model was partially pinned to reserved RAM: 56 large blocks spread across 13183.16 MB
Hooked to model 'transformer' (X0Model)
Async loading plan for model 'transformer' : base size of 643.22 MB will be preloaded with a 737.52 MB async circular shuttle
Partial pinning of data of 'text_encoder' to reserved RAM
The model was partially pinned to reserved RAM: 48 large blocks spread across 10400.74 MB
Hooked to model 'text_encoder' (Gemma4UnifiedForCausalLM)
Async loading plan for model 'text_encoder' : base size of 1920.01 MB will be preloaded with a 230.75 MB async circular shuttle
Unable to apply Partial of 'text_embedding_projection' as no isolated main structures were found
Hooked to model 'text_embedding_projection' (GemmaFeaturesExtractorProjLinear)
Partial pinning of data of 'vae' to reserved RAM
The model was partially pinned to reserved RAM: 4 large blocks spread across 769.54 MB
Hooked to model 'vae' (VideoDecoder)
Partial pinning of data of 'video_encoder' to reserved RAM
The model was partially pinned to reserved RAM: 3 large blocks spread across 601.20 MB
Hooked to model 'video_encoder' (VideoEncoder)
Unable to apply Partial of 'audio_encoder' as no isolated main structures were found
Hooked to model 'audio_encoder' (AudioEncoder)
Unable to apply Partial of 'audio_decoder' as no isolated main structures were found
Hooked to model 'audio_decoder' (AudioDecoder)
Partial pinning of data of 'vocoder' to reserved RAM
The model was partially pinned to reserved RAM: 1 large blocks spread across 241.67 MB
Hooked to model 'vocoder' (VocoderWithBWE)
Unable to apply Partial of 'spatial_upsampler' as no isolated main structures were found
Hooked to model 'spatial_upsampler' (LatentUpsampler)
Partial pinning of data of 'video_embeddings_connector' to reserved RAM
The model was partially pinned to reserved RAM: 6 large blocks spread across 864.96 MB
Hooked to model 'video_embeddings_connector' (Embeddings1DConnector)
Async loading plan for model 'video_embeddings_connector' : 1922.68 MB will be preloaded (base size of 1.00 MB + 62.5% of recurrent layers data) with a 384.34 MB async circular shuttle
Unable to pin data of 'audio_embeddings_connector' to reserved RAM as there is no reserved RAM left. Transfer speed from RAM to VRAM may be slower.        
Hooked to model 'audio_embeddings_connector' (Embeddings1DConnector)
  0%|                                                                                                                            | 0/8 [00:00<?, ?steps/s]NVFP4: linear fallback
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [01:01<00:00,  7.74s/steps]
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [00:58<00:00, 19.51s/steps]
MIOpen Error: D:/jam/TheRock/ml-libs/MIOpen/src/ocl/convolutionocl.cpp:536: Requested convolution is not supported or Immediate mode Fallback unsuccessful.
Traceback (most recent call last):
  File "D:\pinokio\api\wan.git\app\wgp.py", line 7735, in generate_media
    samples = wan_model.generate(
              ^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\models\ltx2\ltx2.py", line 1827, in generate
    pipeline_output = run_ltx2_pipeline(
                      ^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\models\ltx2\ltx2.py", line 1757, in run_ltx2_pipeline
    return self.pipeline(**pipeline_kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\models\ltx2\ltx_pipelines\distilled.py", line 1025, in __call__
    decoded_video = vae_decode_video_to_tensor(
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\models\ltx2\ltx_core\model\video_vae\video_vae.py", line 1106, in decode_video_to_tensor
    for frames in tiled_iterator:
  File "D:\pinokio\api\wan.git\app\models\ltx2\ltx_core\model\video_vae\video_vae.py", line 746, in tiled_decode
    curr_weights = self._accumulate_temporal_group_into_buffer(
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\models\ltx2\ltx_core\model\video_vae\video_vae.py", line 907, in _accumulate_temporal_group_into_buffer
    decoded_tile = self.forward(latent[tile.in_coords], timestep, generator)
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\mmgp\offload.py", line 3359, in check_change_module
    return previous_method(*args, **kwargs)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\models\ltx2\ltx_core\model\video_vae\video_vae.py", line 585, in forward
    sample = up_block(sample, **block_kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\torch\nn\modules\module.py", line 1762, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\mmgp\offload.py", line 3337, in check_load_into_GPU_needed_other
    return previous_method(*args, **kwargs) # other
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\models\ltx2\ltx_core\model\video_vae\resnet.py", line 267, in forward
    hidden_states = resnet(
                    ^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\torch\nn\modules\module.py", line 1762, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\mmgp\offload.py", line 3337, in check_load_into_GPU_needed_other
    return previous_method(*args, **kwargs) # other
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\models\ltx2\ltx_core\model\video_vae\resnet.py", line 149, in forward
    hidden_states = self.conv1(hidden_states, causal=causal)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\torch\nn\modules\module.py", line 1762, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\mmgp\offload.py", line 3337, in check_load_into_GPU_needed_other
    return previous_method(*args, **kwargs) # other
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\models\ltx2\ltx_core\model\video_vae\convolution.py", line 312, in forward
    x = self.conv(x)
        ^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\torch\nn\modules\module.py", line 1751, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\torch\nn\modules\module.py", line 1762, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\mmgp\offload.py", line 3337, in check_load_into_GPU_needed_other
    return previous_method(*args, **kwargs) # other
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\torch\nn\modules\conv.py", line 725, in forward
    return self._conv_forward(input, self.weight, self.bias)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\torch\nn\modules\conv.py", line 709, in _conv_forward
    return F.conv3d(
           ^^^^^^^^^
  File "D:\pinokio\api\wan.git\app\venv\Lib\site-packages\torch\utils\_device.py", line 104, in __torch_function__
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
RuntimeError: miopenStatusNotImplemented. This error may appear if you passed in a non-contiguous input.
Error Queue autosaved successfully to error_queue.zip

Replies (0)

Up to 10 files, 25MB each. Images are optimized; GIFs -> MP4; videos 720p (max 120s).