Skip to content

Actions: NVIDIA/TransformerEngine

Actions

Documentation

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
docs.yml
2,500+ workflow runs
2,500+ workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
GroupedTensor Machinery for MXFP8 Dispatch input passed from Megatron Documentation #21253: Pull request #3660 opened by vthumbe1503
1m 26s vthumbe1503:gt_support_for_mxfp8_nccl
vthumbe1503:gt_support_for_mxfp8_nccl
1m 26s
Abstract CUDA hardcodes into configurable te_device_type / te_platform Documentation #21252: Pull request #3513 synchronize by lxd-cumt
Action required lxd-cumt:cuda_patch_main_latest
lxd-cumt:cuda_patch_main_latest
Action required
[Common] More CuTeDSL MXFP8 specialized kernels Documentation #21251: Pull request #3642 synchronize by pre-commit-ci Bot
1m 21s kainzhong:more_cutedsl_mxfp8_specialized
kainzhong:more_cutedsl_mxfp8_specialized
1m 21s
[Common] More CuTeDSL MXFP8 specialized kernels Documentation #21250: Pull request #3642 synchronize by kainzhong
1m 33s kainzhong:more_cutedsl_mxfp8_specialized
kainzhong:more_cutedsl_mxfp8_specialized
1m 33s
Let the CUDA allocator manage cuBLAS workspaces Documentation #21249: Pull request #3625 synchronize by Vtmpas
Action required Vtmpas:fix/cublas-workspace-per-stream
Vtmpas:fix/cublas-workspace-per-stream
Action required
[Common] Fuse RMSNorm forward with MXFP8 quantization Documentation #21248: Pull request #3657 synchronize by vthumbe1503
1m 42s sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
1m 42s
Fix FP8 current scaling FakeTensor output shape Documentation #21247: Pull request #3659 synchronize by sanjana658
-1s sanjana658:fix/3636-fp8-fake-scale-shape
sanjana658:fix/3636-fp8-fake-scale-shape
-1s
Fix FP8 current scaling FakeTensor output shape Documentation #21246: Pull request #3659 synchronize by sanjana658
Action required sanjana658:fix/3636-fp8-fake-scale-shape
sanjana658:fix/3636-fp8-fake-scale-shape
Action required
Fix FP8 current scaling FakeTensor output shape Documentation #21245: Pull request #3659 opened by sanjana658
Action required sanjana658:fix/3636-fp8-fake-scale-shape
sanjana658:fix/3636-fp8-fake-scale-shape
Action required
[PyTorch] Distributed tests set device in NCCL init and manually shut down NCCL Documentation #21244: Pull request #3655 synchronize by timmoon10
1m 39s timmoon10:tmoon/debug-dist-tests
timmoon10:tmoon/debug-dist-tests
1m 39s
[PyTorch] Deprecate grouped linear and MLP environment gates Documentation #21243: Pull request #3584 synchronize by zhongbozhu
1m 33s zhongbozhu:cleanup/grouped-mlp-env-gates
zhongbozhu:cleanup/grouped-mlp-env-gates
1m 33s
[PyTorch] Distributed tests set device in NCCL init and manually shut down NCCL Documentation #21242: Pull request #3655 synchronize by timmoon10
1m 9s timmoon10:tmoon/debug-dist-tests
timmoon10:tmoon/debug-dist-tests
1m 9s
[Common] Fuse RMSNorm forward with MXFP8 quantization Documentation #21241: Pull request #3657 synchronize by pre-commit-ci Bot
1m 30s sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
1m 30s
[Common] Fuse RMSNorm forward with MXFP8 quantization Documentation #21240: Pull request #3657 synchronize by timmoon10
1m 57s sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
1m 57s
[Common] Fuse RMSNorm forward with MXFP8 quantization Documentation #21239: Pull request #3657 synchronize by timmoon10
1m 33s sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
1m 33s
Add DSv4 cudnn sparse attention kernel adapters to TE Documentation #21238: Pull request #3566 synchronize by sudhakarsingh27
1m 34s sudhakarsingh27:sudhakars/dsv4-cudnn-wrapper
sudhakarsingh27:sudhakars/dsv4-cudnn-wrapper
1m 34s
[Common] Fuse RMSNorm forward with MXFP8 quantization Documentation #21237: Pull request #3657 synchronize by sraman-rgb
1m 8s sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
1m 8s
feat(attention): cuDNN FROST attention backend for head_dim in (256, 512], with context parallelism Documentation #21236: Pull request #3527 synchronize by nvegesna-netizen
1m 38s nvegesna-netizen:nvegesna/te-frost-d512-cp
nvegesna-netizen:nvegesna/te-frost-d512-cp
1m 38s
[Common] Fuse RMSNorm forward with MXFP8 quantization Documentation #21235: Pull request #3657 synchronize by sraman-rgb
1m 38s sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
sraman-rgb:sraman/rmsnorm-mxfp8-fused-kernel
1m 38s
feat(attention): cuDNN FROST attention backend for head_dim in (256, 512], with context parallelism Documentation #21234: Pull request #3527 synchronize by nvegesna-netizen
1m 36s nvegesna-netizen:nvegesna/te-frost-d512-cp
nvegesna-netizen:nvegesna/te-frost-d512-cp
1m 36s
feat(attention): cuDNN FROST attention backend for head_dim in (256, 512], with context parallelism Documentation #21233: Pull request #3527 synchronize by nvegesna-netizen
1m 37s nvegesna-netizen:nvegesna/te-frost-d512-cp
nvegesna-netizen:nvegesna/te-frost-d512-cp
1m 37s
feat(attention): cuDNN FROST attention backend for head_dim in (256, 512], with context parallelism Documentation #21232: Pull request #3527 synchronize by nvegesna-netizen
1m 29s nvegesna-netizen:nvegesna/te-frost-d512-cp
nvegesna-netizen:nvegesna/te-frost-d512-cp
1m 29s
feat(attention): cuDNN FROST attention backend for head_dim in (256, 512], with context parallelism Documentation #21231: Pull request #3527 synchronize by nvegesna-netizen
1m 49s nvegesna-netizen:nvegesna/te-frost-d512-cp
nvegesna-netizen:nvegesna/te-frost-d512-cp
1m 49s
feat(attention): cuDNN FROST attention backend for head_dim in (256, 512], with context parallelism Documentation #21230: Pull request #3527 synchronize by pre-commit-ci Bot
1m 44s nvegesna-netizen:nvegesna/te-frost-d512-cp
nvegesna-netizen:nvegesna/te-frost-d512-cp
1m 44s
Let the CUDA allocator manage cuBLAS workspaces Documentation #21229: Pull request #3625 synchronize by Vtmpas
Action required Vtmpas:fix/cublas-workspace-per-stream
Vtmpas:fix/cublas-workspace-per-stream
Action required