ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
unsloth/studio/backend/utils/hardware
History
Daniel Han cc6c8160d1 Add IS_ROCM flag to hardware module and fix AMD error message 
Hardware module changes:
- Add IS_ROCM global flag (bool, default False) to hardware.py
- detect_hardware() sets IS_ROCM=True when torch.version.hip is set,
  while keeping DeviceType as CUDA (torch.cuda.* works on ROCm via HIP)
- Print "ROCm (HIP X.Y)" instead of just "CUDA" when IS_ROCM is True
- get_package_versions() now returns a "rocm" key with the HIP version
  (torch.version.hip) alongside the existing "cuda" key
- Export IS_ROCM from studio/backend/utils/hardware/__init__.py

Tokenizer error message fix:
- Replace "We do not support AMD" with a helpful message pointing to
  ROCm installation docs at docs.unsloth.ai, since AMD is now supported
2026-03-31 08:34:05 +00:00
..
__init__.py Add IS_ROCM flag to hardware module and fix AMD error message 2026-03-31 08:34:05 +00:00
hardware.py Add IS_ROCM flag to hardware module and fix AMD error message 2026-03-31 08:34:05 +00:00
nvidia.py [Studio] multi gpu finetuning/inference via "balanced_low0/sequential" device_map (#4602) 2026-03-30 02:33:15 -07:00
VRAM_ESTIMATION.md [Studio] multi gpu finetuning/inference via "balanced_low0/sequential" device_map (#4602) 2026-03-30 02:33:15 -07:00
vram_estimation.py [Studio] multi gpu finetuning/inference via "balanced_low0/sequential" device_map (#4602) 2026-03-30 02:33:15 -07:00