pytorch/xla

Pull Requests Commits

Bump CUDNN version in docker image. (#8718)

ysiraichi committed 258 days ago

Verified 60160233

Bump XLA commit pin. (#8714)

ysiraichi committed 258 days ago

Verified ca248c7e

Support splitting physical axis in SPMD mesh (#8698)

tengyifei committed 262 days ago

Verified 8a712495

update pin (#8677)

lsy323 committed 263 days ago

Verified 6f423d0b

Fix CUDA plugin CI. (#8593)

ysiraichi committed 264 days ago

Verified 42edbe18

Include the alias ID in GetXLATensorDebugInfo (#8701)

rpsilva-aws committed 264 days ago

Verified 3a4ad6f3

Integrate ragged paged attn to pytorch/xla (#8692)

bythew3i committed 265 days ago

Verified 06e1b591

Mention remote build in CONTRIBUTING.md (#8699)

tengyifei committed 265 days ago

Verified 45af5193

Add missing NeuronPlugin configure_single_process function (#8694)

jeffhataws committed 266 days ago

Verified 065cb5b1

Fix a bug in flash attention where kv_seq_len should divide block_k_major. (#8671)

zhangp365 committed 266 days ago

Verified cff9f4e0

Add another C++11 ABI docker reference (#8649)

tengyifei committed 269 days ago

Verified c0afda3a

[torchax] Update installation guide in readme (#8664)

lsy323 committed 269 days ago

Verified ace5e24b

Fix dk/dv autograd error on TPU flash attention (#8685)

zmelumian972 committed 269 days ago

Verified 0cd1fc2a

Added pow to autocast policy and unit test (#8684)

avizon-aws committed 269 days ago

Verified afaf0d05

Misc fixes (#8679)

qihqi committed 270 days ago

Verified 8b45e599

Fix aten.linear and add aten.einsum (#8666)

yaochengji committed 270 days ago

Verified 45109d1e

fixed the type miss match issue with symint in SPMD regime and dynamo (#8681)

hosseinsarshar committed 270 days ago

Verified b9bef193

fix a bug in tensor.flatten (#8680)

yaochengji committed 272 days ago

Verified 49cd3105

Install `clang-17` on development Dockerfile. (#8673)

ysiraichi committed 272 days ago

Verified ec10537c

[CI] Fix torchbench setup failure (#8670)

lsy323 committed 272 days ago

Verified a4356250

Add ragged paged attention (#8659)

vanbasten23 committed 273 days ago

Verified 84800944

`cummax`: fix 0-sized dimension reduction. (#8653)

ysiraichi committed 273 days ago

Verified 1c537904

[torchax][distributed] Use local device for default jax device (#8667)

lsy323 committed 273 days ago

Verified 35789406

Dynamo/AOTAutograd traceable flash attention (#8654)

zpcore committed 275 days ago

Verified 9ae017e2

openxla pin update to 20250131 (#8621)

zpcore committed 276 days ago

Verified 82d35047

[torchax] fix einsum op conversion (#8657)

yaochengji committed 276 days ago

Verified 8572e750

Fix the retention for existing gradients in the Grad Acc API (#8658)

rpsilva-aws committed 276 days ago

Verified 39dd7959

Add a stub for host offloading docs (#8656)

tengyifei committed 277 days ago

Verified e583c2cc

Fix C++11 ABI wheel instructions (#8650)

tengyifei committed 278 days ago

Verified 93a2ba6b

Update README.md for PyTorch/XLA 2.6 release (#8647)

tengyifei committed 278 days ago

Verified d9593f03

Newer Older