Running TRELLIS.2 on A100 gives cuda error

[HrsPythonix]

cmd:

docker stop tostui-trellis2; docker rm tostui-trellis2; docker pull camenduru/tostui-trellis2
docker run --gpus all -p 3000:3000 --name tostui-trellis2 camenduru/tostui-trellis2

Log:

INFO   | test-47da1e5e-4a7d-44e9-88c8-1cbb489cff16 | Started.
Sampling sparse structure: 100%|██████████| 12/12 [00:02<00:00,  5.11it/s]
Sampling shape SLat: 100%|██████████| 12/12 [00:01<00:00,  8.24it/s]
Directory /content/TRELLIS.2/tmp has been removed successfully.
DEBUG  | test-47da1e5e-4a7d-44e9-88c8-1cbb489cff16 | Handler output: {'job_id': 'efa63586a60c6efd7ef82c73b7d4b5c4', 'result': 'CUDA error: no kernel image is available for execution on the device\nCUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.\nFor debugging consider passing CUDA_LAUNCH_BLOCKING=1\nCompile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.\n', 'status': 'FAILED'}
DEBUG  | test-47da1e5e-4a7d-44e9-88c8-1cbb489cff16 | run_job return: {'output': {'job_id': 'efa63586a60c6efd7ef82c73b7d4b5c4', 'result': 'CUDA error: no kernel image is available for execution on the device\nCUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.\nFor debugging consider passing CUDA_LAUNCH_BLOCKING=1\nCompile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.\n', 'status': 'FAILED'}}
Image proxy error: TypeError: Failed to parse URL from CUDA error: no kernel image is available for execution on the device
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

    at d (.next/server/chunks/ssr/[root-of-the-server]__6b142534._.js:4:26849)
    at R (.next/server/chunks/ssr/[root-of-the-server]__6b142534._.js:4:33062)
    at <unknown> (.next/server/chunks/ssr/[root-of-the-server]__6b142534._.js:4:35523)
    at <unknown> (.next/server/chunks/ssr/e559b_next_dist_66d926d9._.js:1:35478)
    at <unknown> (.next/server/chunks/ssr/e559b_next_dist_66d926d9._.js:1:35023) {
  [cause]: TypeError: Invalid URL
      at d (.next/server/chunks/ssr/[root-of-the-server]__6b142534._.js:4:26849)
      at R (.next/server/chunks/ssr/[root-of-the-server]__6b142534._.js:4:33062)
      at <unknown> (.next/server/chunks/ssr/[root-of-the-server]__6b142534._.js:4:35523)
      at <unknown> (.next/server/chunks/ssr/e559b_next_dist_66d926d9._.js:1:35478) {
    code: 'ERR_INVALID_URL',
    input: 'CUDA error: no kernel image is available for execution on the device\n' +
      'CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.\n' +
      'For debugging consider passing CUDA_LAUNCH_BLOCKING=1\n' +
      'Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.\n'

[camenduru]

I built the libraries with TORCH_CUDA_ARCH_LIST=8.6;8.9;9.0;9.0a;10.0;12.0

A100: 8.0 😐 I'll add A100 in the next update.

[zjy200809]

看到昨天更新，可以兼容A100: 8.0了吗？