Setup error

[nick-morhun]

Operating System: Ubuntu 24.04.3 LTS
Kernel: Linux 6.6.87.2-microsoft-standard-WSL2
Architecture: x86-64

I am trying to run on Ubuntu: pip install --extra-index-url https://download.pytorch.org/whl/cu128 -e ".[flash-attn]".
This is what I get:

Looking in indexes: https://pypi.org/simple, https://download.pytorch.org/whl/cu128
Obtaining file:///home/nickmorhun/MOSS-TTS
  Installing build dependencies ... done
  Checking if build backend supports build_editable ... done
  Getting requirements to build editable ... done
  Preparing editable metadata (pyproject.toml) ... done
Collecting torch==2.9.1+cu128 (from moss-tts==0.1.0)
  Using cached https://download.pytorch.org/whl/cu128/torch-2.9.1%2Bcu128-cp312-cp312-manylinux_2_28_x86_64.whl.metadata (30 kB)
Collecting torchaudio==2.9.1+cu128 (from moss-tts==0.1.0)
  Using cached https://download.pytorch.org/whl/cu128/torchaudio-2.9.1%2Bcu128-cp312-cp312-manylinux_2_28_x86_64.whl.metadata (6.9 kB)
Collecting torchcodec===0.8.1 (from moss-tts==0.1.0)
  Using cached torchcodec-0.8.1-cp312-cp312-manylinux_2_28_x86_64.whl.metadata (9.7 kB)
Collecting transformers==5.0.0 (from moss-tts==0.1.0)
  Using cached transformers-5.0.0-py3-none-any.whl.metadata (37 kB)
Collecting safetensors==0.6.2 (from moss-tts==0.1.0)
  Using cached safetensors-0.6.2-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (4.1 kB)
Collecting numpy==2.1.0 (from moss-tts==0.1.0)
  Using cached numpy-2.1.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (60 kB)
Collecting orjson==3.11.4 (from moss-tts==0.1.0)
  Using cached orjson-3.11.4-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (41 kB)
Collecting tqdm==4.67.1 (from moss-tts==0.1.0)
  Using cached tqdm-4.67.1-py3-none-any.whl.metadata (57 kB)
Collecting PyYAML==6.0.3 (from moss-tts==0.1.0)
  Using cached pyyaml-6.0.3-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl.metadata (2.4 kB)
Collecting einops==0.8.1 (from moss-tts==0.1.0)
  Using cached einops-0.8.1-py3-none-any.whl.metadata (13 kB)
Collecting scipy==1.16.2 (from moss-tts==0.1.0)
  Using cached scipy-1.16.2-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (62 kB)
Collecting librosa==0.11.0 (from moss-tts==0.1.0)
  Using cached librosa-0.11.0-py3-none-any.whl.metadata (8.7 kB)
Collecting tiktoken==0.12.0 (from moss-tts==0.1.0)
  Using cached tiktoken-0.12.0-cp312-cp312-manylinux_2_28_x86_64.whl.metadata (6.7 kB)
Collecting psutil (from moss-tts==0.1.0)
  Using cached psutil-7.2.2-cp36-abi3-manylinux2010_x86_64.manylinux_2_12_x86_64.manylinux_2_28_x86_64.whl.metadata (22 kB)
Collecting packaging (from moss-tts==0.1.0)
  Using cached packaging-26.0-py3-none-any.whl.metadata (3.3 kB)
Collecting ninja (from moss-tts==0.1.0)
  Using cached ninja-1.13.0-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (5.1 kB)
Collecting setuptools (from moss-tts==0.1.0)
  Using cached setuptools-82.0.0-py3-none-any.whl.metadata (6.6 kB)
Collecting wheel (from moss-tts==0.1.0)
  Using cached wheel-0.46.3-py3-none-any.whl.metadata (2.4 kB)
Collecting gradio (from moss-tts==0.1.0)
  Using cached gradio-6.5.1-py3-none-any.whl.metadata (16 kB)
Collecting audioread>=2.1.9 (from librosa==0.11.0->moss-tts==0.1.0)
  Using cached audioread-3.1.0-py3-none-any.whl.metadata (9.0 kB)
Collecting numba>=0.51.0 (from librosa==0.11.0->moss-tts==0.1.0)
  Using cached numba-0.63.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (2.9 kB)
Collecting scikit-learn>=1.1.0 (from librosa==0.11.0->moss-tts==0.1.0)
  Using cached scikit_learn-1.8.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (11 kB)
Collecting joblib>=1.0 (from librosa==0.11.0->moss-tts==0.1.0)
  Using cached joblib-1.5.3-py3-none-any.whl.metadata (5.5 kB)
Collecting decorator>=4.3.0 (from librosa==0.11.0->moss-tts==0.1.0)
  Using cached decorator-5.2.1-py3-none-any.whl.metadata (3.9 kB)
Collecting soundfile>=0.12.1 (from librosa==0.11.0->moss-tts==0.1.0)
  Using cached soundfile-0.13.1-py2.py3-none-manylinux_2_28_x86_64.whl.metadata (16 kB)
Collecting pooch>=1.1 (from librosa==0.11.0->moss-tts==0.1.0)
  Using cached pooch-1.9.0-py3-none-any.whl.metadata (10 kB)
Collecting soxr>=0.3.2 (from librosa==0.11.0->moss-tts==0.1.0)
  Using cached soxr-1.0.0-cp312-abi3-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (5.6 kB)
Collecting typing_extensions>=4.1.1 (from librosa==0.11.0->moss-tts==0.1.0)
  Using cached https://download.pytorch.org/whl/typing_extensions-4.15.0-py3-none-any.whl.metadata (3.3 kB)
Collecting lazy_loader>=0.1 (from librosa==0.11.0->moss-tts==0.1.0)
  Using cached lazy_loader-0.4-py3-none-any.whl.metadata (7.6 kB)
Collecting msgpack>=1.0 (from librosa==0.11.0->moss-tts==0.1.0)
  Using cached msgpack-1.1.2-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl.metadata (8.1 kB)
Collecting regex>=2022.1.18 (from tiktoken==0.12.0->moss-tts==0.1.0)
  Using cached regex-2026.1.15-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl.metadata (40 kB)
Collecting requests>=2.26.0 (from tiktoken==0.12.0->moss-tts==0.1.0)
  Using cached requests-2.32.5-py3-none-any.whl.metadata (4.9 kB)
Collecting filelock (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Using cached filelock-3.24.2-py3-none-any.whl.metadata (2.0 kB)
Collecting sympy>=1.13.3 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Using cached sympy-1.14.0-py3-none-any.whl.metadata (12 kB)
Collecting networkx>=2.5.1 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Using cached networkx-3.6.1-py3-none-any.whl.metadata (6.8 kB)
Collecting jinja2 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Using cached https://download.pytorch.org/whl/jinja2-3.1.6-py3-none-any.whl.metadata (2.9 kB)
Collecting fsspec>=0.8.5 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Using cached fsspec-2026.2.0-py3-none-any.whl.metadata (10 kB)
Collecting nvidia-cuda-nvrtc-cu12==12.8.93 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-cuda-nvrtc-cu12/nvidia_cuda_nvrtc_cu12-12.8.93-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl (88.0 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 88.0/88.0 MB 6.7 MB/s eta 0:00:00
Collecting nvidia-cuda-runtime-cu12==12.8.90 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-cuda-runtime-cu12/nvidia_cuda_runtime_cu12-12.8.90-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (954 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 954.8/954.8 kB 7.5 MB/s eta 0:00:00
Collecting nvidia-cuda-cupti-cu12==12.8.90 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-cuda-cupti-cu12/nvidia_cuda_cupti_cu12-12.8.90-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (10.2 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 10.2/10.2 MB 6.3 MB/s eta 0:00:00
Collecting nvidia-cudnn-cu12==9.10.2.21 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-cudnn-cu12/nvidia_cudnn_cu12-9.10.2.21-py3-none-manylinux_2_27_x86_64.whl (706.8 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 706.8/706.8 MB 6.5 MB/s eta 0:00:00
Collecting nvidia-cublas-cu12==12.8.4.1 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-cublas-cu12/nvidia_cublas_cu12-12.8.4.1-py3-none-manylinux_2_27_x86_64.whl (594.3 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 594.3/594.3 MB 6.5 MB/s eta 0:00:00
Collecting nvidia-cufft-cu12==11.3.3.83 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-cufft-cu12/nvidia_cufft_cu12-11.3.3.83-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (193.1 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 193.1/193.1 MB 6.6 MB/s eta 0:00:00
Collecting nvidia-curand-cu12==10.3.9.90 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-curand-cu12/nvidia_curand_cu12-10.3.9.90-py3-none-manylinux_2_27_x86_64.whl (63.6 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 63.6/63.6 MB 6.4 MB/s eta 0:00:00
Collecting nvidia-cusolver-cu12==11.7.3.90 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-cusolver-cu12/nvidia_cusolver_cu12-11.7.3.90-py3-none-manylinux_2_27_x86_64.whl (267.5 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 267.5/267.5 MB 6.6 MB/s eta 0:00:00
Collecting nvidia-cusparse-cu12==12.5.8.93 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-cusparse-cu12/nvidia_cusparse_cu12-12.5.8.93-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (288.2 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 288.2/288.2 MB 6.6 MB/s eta 0:00:00
Collecting nvidia-cusparselt-cu12==0.7.1 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-cusparselt-cu12/nvidia_cusparselt_cu12-0.7.1-py3-none-manylinux2014_x86_64.whl (287.2 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 287.2/287.2 MB 6.6 MB/s eta 0:00:00
Collecting nvidia-nccl-cu12==2.27.5 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-nccl-cu12/nvidia_nccl_cu12-2.27.5-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (322.3 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 322.3/322.3 MB 5.2 MB/s eta 0:00:00
Collecting nvidia-nvshmem-cu12==3.3.20 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-nvshmem-cu12/nvidia_nvshmem_cu12-3.3.20-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (124.7 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 124.7/124.7 MB 5.8 MB/s eta 0:00:00
Collecting nvidia-nvtx-cu12==12.8.90 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-nvtx-cu12/nvidia_nvtx_cu12-12.8.90-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (89 kB)
Collecting nvidia-nvjitlink-cu12==12.8.93 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-nvjitlink-cu12/nvidia_nvjitlink_cu12-12.8.93-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl (39.3 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 39.3/39.3 MB 5.8 MB/s eta 0:00:00
Collecting nvidia-cufile-cu12==1.13.1.3 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Downloading https://pypi.nvidia.com/nvidia-cufile-cu12/nvidia_cufile_cu12-1.13.1.3-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (1.2 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.2/1.2 MB 6.8 MB/s eta 0:00:00
Collecting triton==3.5.1 (from torch==2.9.1+cu128->moss-tts==0.1.0)
  Using cached https://download.pytorch.org/whl/triton-3.5.1-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (1.7 kB)
Collecting huggingface-hub<2.0,>=1.3.0 (from transformers==5.0.0->moss-tts==0.1.0)
  Using cached huggingface_hub-1.4.1-py3-none-any.whl.metadata (13 kB)
Collecting tokenizers<=0.23.0,>=0.22.0 (from transformers==5.0.0->moss-tts==0.1.0)
  Using cached tokenizers-0.22.2-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (7.3 kB)
Collecting typer-slim (from transformers==5.0.0->moss-tts==0.1.0)
  Using cached typer_slim-0.23.1-py3-none-any.whl.metadata (4.2 kB)
Collecting flash-attn (from moss-tts==0.1.0)
  Using cached flash_attn-2.8.3.tar.gz (8.4 MB)
  Installing build dependencies ... done
  Getting requirements to build wheel ... error
  error: subprocess-exited-with-error

  × Getting requirements to build wheel did not run successfully.
  │ exit code: 1
  ╰─> [22 lines of output]
      /tmp/pip-build-env-jxjqv64f/overlay/lib/python3.12/site-packages/setuptools/_vendor/wheel/bdist_wheel.py:4: FutureWarning: The 'wheel' package is no longer the canonical location of the 'bdist_wheel' command, and will be removed in a future release. Please update to setuptools v70.1 or later which contains an integrated version of this command.
        warn(
      Traceback (most recent call last):
        File "/home/nickmorhun/.pyenv/versions/3.12.12/envs/mossTTSenv/lib/python3.12/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 389, in <module>
          main()
        File "/home/nickmorhun/.pyenv/versions/3.12.12/envs/mossTTSenv/lib/python3.12/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 373, in main
          json_out["return_val"] = hook(**hook_input["kwargs"])
                                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        File "/home/nickmorhun/.pyenv/versions/3.12.12/envs/mossTTSenv/lib/python3.12/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 143, in get_requires_for_build_wheel
          return hook(config_settings)
                 ^^^^^^^^^^^^^^^^^^^^^
        File "/tmp/pip-build-env-jxjqv64f/overlay/lib/python3.12/site-packages/setuptools/build_meta.py", line 333, in get_requires_for_build_wheel
          return self._get_build_requires(config_settings, requirements=[])
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        File "/tmp/pip-build-env-jxjqv64f/overlay/lib/python3.12/site-packages/setuptools/build_meta.py", line 301, in _get_build_requires
          self.run_setup()
        File "/tmp/pip-build-env-jxjqv64f/overlay/lib/python3.12/site-packages/setuptools/build_meta.py", line 520, in run_setup
          super().run_setup(setup_script=setup_script)
        File "/tmp/pip-build-env-jxjqv64f/overlay/lib/python3.12/site-packages/setuptools/build_meta.py", line 317, in run_setup
          exec(code, locals())
        File "<string>", line 22, in <module>
      ModuleNotFoundError: No module named 'torch'
      [end of output]

  note: This error originates from a subprocess, and is likely not a problem with pip.

[xiami2019]

You need to install torch at first then FA2.
We will fix our README to make it easier to setup.

[mateuspestana]

I'm experiencing the same, but I already have torch installed. How to fix this? My GPU is RTX 3060 12gb. I'm on Omarchy.