Skip to content

Actions: NVIDIA/TransformerEngine

Actions

TE-CI Trigger

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
trigger-ci.yml
3,856 workflow runs
3,856 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

[Common] Enable determinism for cuDNN >= 9.18.1 on Blackwell TE-CI Trigger #8157: Issue comment #2584 (comment) created by cyanguwa
1s
1s
[Common] Enable determinism for cuDNN >= 9.18.1 on Blackwell TE-CI Trigger #8156: Issue comment #2584 (comment) created by cyanguwa
3m 32s
3m 32s
Add support for SWA (left, right) with FusedAttention TE-CI Trigger #8155: Issue comment #2477 (comment) created by cyanguwa
1s
1s
fused_attn_fwd and fused_attn_bwd take many time on cpu TE-CI Trigger #8154: Issue comment #2033 (comment) created by FuyuanChen
1s
1s
Build Improvements - Find NCCL.h automatically from pypi nvidia-nccl-cu12/cu13 TE-CI Trigger #8153: Issue comment #2331 (comment) created by yangfanww
2s
2s
Why NVFP4 Inference (50 PFLOPS) Outperforms Training (35 PFLOPS) on Rubin GPU？ TE-CI Trigger #8152: Issue comment #2565 (comment) created by YangWang92
1s
1s
Add logic for block-scaled tensors with GEMM swizzled scales TE-CI Trigger #8151: Issue comment #2486 (comment) created by timmoon10
3m 21s
3m 21s
TE-CI Trigger TE-CI Trigger #8150: created by timmoon10
31s
31s
CPU Optimizations for FP8 TE-CI Trigger #8149: Issue comment #2559 (comment) created by vthumbe1503
3m 28s
3m 28s
[Common] Enable determinism for cuDNN >= 9.18.1 on Blackwell TE-CI Trigger #8148: Issue comment #2584 (comment) created by KshitijLakhani
3m 26s
3m 26s
[Common] Tuned NVFP4 cast kernel TE-CI Trigger #8147: Issue comment #2412 (comment) created by Oleg-Goncharov
3m 27s
3m 27s
[Common] Tuned NVFP4 cast kernel TE-CI Trigger #8146: Issue comment #2412 (comment) created by greptile-apps bot
1s
1s
[pyTorch] CPU performance optimizations TE-CI Trigger #8145: Issue comment #2439 (comment) created by ptrendx
3m 20s
3m 20s
[common] Add support for cuBLASLt GEMM for GroupedTensor TE-CI Trigger #8144: Issue comment #2502 (comment) created by ptrendx
3m 32s
3m 32s
Add support for SWA (left, right) with FusedAttention TE-CI Trigger #8143: Issue comment #2477 (comment) created by sudhakarsingh27
3m 16s
3m 16s
fix(examples): te_llama compatibility with transformers >= 4.57 TE-CI Trigger #8142: Issue comment #2572 (comment) created by sudhakarsingh27
1s
1s
[JAX] Custom partitioning for Permutation primitives TE-CI Trigger #8141: Issue comment #2591 (comment) created by tdophung
3m 36s
3m 36s
[Common] Enable determinism for cuDNN >= 9.18.1 on Blackwell TE-CI Trigger #8140: Issue comment #2584 (comment) created by liayan
2s
2s
Add logic for block-scaled tensors with GEMM swizzled scales TE-CI Trigger #8139: Issue comment #2486 (comment) created by timmoon10
3m 31s
3m 31s
Is the version constraint on flash-attn real? TE-CI Trigger #8138: Issue comment #519 (comment) created by sbhavani
1s
1s
fp8 is slower than fp16 TE-CI Trigger #8137: Issue comment #445 (comment) created by sbhavani
4s
4s
[Feature Request] Please support pytorch and Jax FP8 type without custom data type TE-CI Trigger #8136: Issue comment #391 (comment) created by sbhavani
1s
1s
FP8 Memory Usage TE-CI Trigger #8135: Issue comment #257 (comment) created by sbhavani
1s
1s
[PyTorch] Support user-defined op fusions TE-CI Trigger #8134: Issue comment #2597 (comment) created by timmoon10
3m 32s
3m 32s
[PyTorch] Support user-defined op fusions TE-CI Trigger #8133: Issue comment #2597 (comment) created by timmoon10
3m 37s
3m 37s