ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
unsloth/studio
History
Roland Tannous ce28dbd4c1 fix: derive HF_HUB_CACHE from HF_HOME when set 
Previously HF_HUB_CACHE always defaulted to ~/.cache/huggingface/hub
even when HF_HOME was explicitly set (e.g. in Docker). This caused
models to download to the wrong location instead of the configured
HF_HOME path.
2026-04-05 05:22:43 +00:00
..
backend fix: derive HF_HUB_CACHE from HF_HOME when set 2026-04-05 05:22:43 +00:00
frontend studio: add speculative decoding support (ngram-mod, on by default) (#4836) 2026-04-03 13:56:59 -07:00
__init__.py Final cleanup 2026-03-12 18:28:04 +00:00
install_llama_prebuilt.py Fix/llama.cppbuilding (#4804) 2026-04-03 00:34:20 -07:00
install_python_stack.py add Docker support: skip venv, install only missing deps 2026-04-04 02:02:26 +00:00
LICENSE.AGPL-3.0 Add AGPL-3.0 license to studio folder 2026-03-09 19:36:25 +00:00
setup.bat Final cleanup 2026-03-12 18:28:04 +00:00
setup.ps1 Fix/llama.cppbuilding (#4804) 2026-04-03 00:34:20 -07:00
setup.sh Skip llama.cpp install in Docker mode 2026-04-05 04:50:43 +00:00
Unsloth_Studio_Colab.ipynb Allow install_python_stack to run on Colab (#4633) 2026-03-27 00:29:27 +04:00