| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| 0.0.76 source code.tar.gz | 2025-03-16 | 2.8 MB | |
| 0.0.76 source code.zip | 2025-03-16 | 2.9 MB | |
| README.md | 2025-03-16 | 944 Bytes | |
| Totals: 3 Items | 5.7 MB | 0 | |
- torch=2.6.0 update - 5-10% faster attention on hopper -> previously 2.4.1 -> does no longer work with torch.compile + bettertransformers. We recommend disabling torch.compile for this model class.
- flash-attn included in docker image for nvidia.
What's Changed
- bump client version by @wirthual in https://github.com/michaelfeil/infinity/pull/522
- add new st version by @michaelfeil in https://github.com/michaelfeil/infinity/pull/523
- Version check step by @wirthual in https://github.com/michaelfeil/infinity/pull/524
- README: add example for using local model wtth docker container by @wirthual in https://github.com/michaelfeil/infinity/pull/528
- add vision client template by @wirthual in https://github.com/michaelfeil/infinity/pull/526
- bump to 2.6 torch by @michaelfeil in https://github.com/michaelfeil/infinity/pull/556
Full Changelog: https://github.com/michaelfeil/infinity/compare/0.0.75...0.0.76