Claude bb6da662de Add AMD ROCm GPU support ...

Introduces a third hardware mode alongside CUDA and CPU: ROCm (AMD GPU).
AMD GPUs present as CUDA devices under PyTorch ROCm, so the existing GPU
path is reused with minimal changes — the main additions are wheel management,
device detection, and suppressing flash_attn (unsupported on ROCm).

- server/vibevoice_server.py: extend _resolve_device() to recognise 'rocm'
  (auto-detected via torch.version.hip); add _torch_device() helper that maps
  'rocm' → 'cuda' for all PyTorch API calls; apply GPU optimisations for both
  cuda and rocm in _init_model(); always use sdpa on ROCm; propagate
  _torch_device() to _load_voice_presets() map_location.
- server/start.sh: add --rocm flag; sync .venv-rocm with uv sync --no-sources
  then replace torch with the ROCm 6.2 wheel via uv pip install; set
  VIBEPOD_DEVICE=rocm for uvicorn.
- server/pyproject.toml: register pytorch-rocm62 index (explicit); add
  .venv-rocm to ruff excludes.
- package.json: add dev:rocm and dev:server:rocm scripts.
- README.md: document ROCm mode, prerequisites (RX 6000+, ROCm 6.2+, Linux),
  and new commands; expand CUDA vs CPU section to CUDA vs CPU vs ROCm.

https://claude.ai/code/session_0168pSswiaoEf6LEx6UQWfBu

2026-05-04 01:54:57 +00:00

..

.python-version

perf: improve streaming generation pipeline

2026-04-30 18:54:14 +01:00

download_model.py

style: apply ruff formatting and lint fixes to server

2026-05-01 19:06:13 +01:00

pyproject.toml

Add AMD ROCm GPU support

2026-05-04 01:54:57 +00:00

start.sh

Add AMD ROCm GPU support

2026-05-04 01:54:57 +00:00

uv.lock

chore: add ruff linter/formatter for Python server

2026-05-01 19:05:39 +01:00

vibevoice_server.py

Add AMD ROCm GPU support

2026-05-04 01:54:57 +00:00