Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
COMMITS
/ .github/workflows November 7, 2025
N
Update ci to min python v3.10 (#21344)
Nicki Skafte Detlefsen committed
November 3, 2025
J
Remove Support For Deprecated Habana (#21327)
Justus Schock committed
October 29, 2025
D
build(deps): bump actions/upload-artifact from 4 to 5 (#21315)
dependabot[bot] committed
D
build(deps): bump actions/download-artifact from 5 to 6 (#21316)
dependabot[bot] committed
October 28, 2025
J
drop dumping wheels after UV switch (#21217)
Jirka Borovec committed
October 15, 2025
D
build(deps): bump astral-sh/setup-uv from 6 to 7 (#21284)
dependabot[bot] committed
September 25, 2025
N
Add note on accessing last and best checkpoint (#21241)
Nicki Skafte Detlefsen committed
September 15, 2025
S
ci: `uv` for `docs-build` (#21206)
Shion Matsumoto committed
B
fix: Checkout and Git LFS fetch issue for pytorch docs workflow (#21219)
Bhimraj Yadav committed
September 10, 2025
J
drop using dockers (#21190)
Jirka Borovec committed
September 8, 2025
J
tests: ignore future warning for oldest configurations (#21185)
Jirka Borovec committed
S
code check with uv (#21183)
Shion Matsumoto committed
J
fix using CPU torch with UV (#21181)
Jirka Borovec committed
D
build(deps): bump actions/labeler from 5 to 6 (#21176)
dependabot[bot] committed
D
build(deps): bump actions/setup-python from 5 to 6 (#21175)
dependabot[bot] committed
September 3, 2025
S
`uv` for tests-fabric (#21155)
Shion Matsumoto committed
S
`uv` for pytorch tests (#21148)
Shion Matsumoto committed
September 2, 2025
N
Add support for deepspeeds `exclude_frozen_parameters` (#21060)
Nicki Skafte Detlefsen committed
S
Fix workflow matrix reference (#21145)
Shion Matsumoto committed
September 1, 2025
S
Simplify Fabric tests workflow matrix (#21142)
Shion Matsumoto committed
S
Simplify workflow matrix (#21132)
Shion Matsumoto committed
D
build(deps): bump google-github-actions/setup-gcloud from 2 to 3 (#21140)
dependabot[bot] committed
D
build(deps): bump google-github-actions/auth from 2 to 3 (#21139)
dependabot[bot] committed
August 21, 2025
J
ci: pin also test requirements for minimal setup (#21102)
Jirka Borovec committed
August 19, 2025
J
switch to lightning_utilities.cli requirements set-oldest (#21077)
Jirka Borovec committed
August 18, 2025
D
build(deps): bump actions/checkout from 4 to 5 (#21091)
dependabot[bot] committed
August 15, 2025
G
Torch-Tensorrt Integration with LightningModule (#20808)
GdoongMathew committed
August 13, 2025
J
chore: bump PyTorch version in dependencies & CI (#21043)
Jirka Borovec committed
August 11, 2025
J
docker: simplify the docker name with CUDA (#21001)
Jirka Borovec committed
D
build(deps): bump actions/download-artifact from 4 to 5 (#21049)
dependabot[bot] committed
D
build(deps): bump Lightning-AI/utilities from 0.15.0 to 0.15.2 (#21050)
dependabot[bot] committed
August 8, 2025
J
docker: build images for latest PT `2.8` (#21042)
Jirka Borovec committed
August 4, 2025
D
build(deps): bump Lightning-AI/utilities from 0.14.3 to 0.15.0 (#21010)
dependabot[bot] committed
July 3, 2025
J
ci: bump sorting group-check (#20955)
Jirka Borovec committed
July 2, 2025
J
ci: force `pip install` with `--upgrade-strategy=eager` (#20958)
Jirka Borovec committed
June 26, 2025
J
ci: disable TPU testing (#20942)
Jirka Borovec committed
June 24, 2025
J
test: addressing flaky spawn "process 0 terminated with signal SIGABRT" (#20933)
Jirka Borovec committed
June 23, 2025
J
fix: update automated checkpoint messages for consistency (#20924)
Jirka Borovec committed
D
build(deps): bump codecov/codecov-action from 4 to 5 (#20927)
dependabot[bot] committed
June 17, 2025
J
ci: debug failing run on master (#20845)
Jirka Borovec committed
J
debugging flaky `test_collective_operations` with SIGABRT (#20912)
Jirka Borovec committed
June 16, 2025
J
fix check for flaky links in readme (#20910)
Jirka Borovec committed
June 5, 2025
J
bump: PyTorch to be latest `2.7.1` (#20877)
Jirka Borovec committed
May 19, 2025
J
docker: update building base docker images for last CUDA & py3.10 (#20844)
Jirka Borovec committed
J
ci: skip failing run on master (#20843)
Jirka Borovec committed
J
docker: extend building base docker images for litGPT (#20842)
Jirka Borovec committed
April 28, 2025
J
drop mergify (#20770)
Jirka Borovec committed
April 25, 2025
J
bump: testing latest PT on GPU to `2.7` (#20754)
Jirka Borovec committed
J
update model message (#20753)
Jirka Borovec committed
April 22, 2025
J
ci: try to supress false failing check for create legacy checkpoint (#20746)
Jirka Borovec committed