Skip to content

Actions: NVIDIA/TransformerEngine

Actions

TE-CI Trigger

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
trigger-ci.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
Quantization support for GroupedTensor: FP8 per-tensor TE-CI Trigger #10309: Issue comment #3102 (comment) created by int-smart
1s
1s
[PyTorch][torch.compile] Decouple amax reduction group from the quantizer TE-CI Trigger #10308: Issue comment #3104 (comment) created by pggPL
3m 50s
3m 50s
[PyTorch][torch.compile] Decouple amax reduction group from the quantizer TE-CI Trigger #10307: Issue comment #3104 (comment) created by pggPL
3s
3s
[Common/PyTorch] Grouped-quantize kernels for 1D and 2D FP8 block-scaling TE-CI Trigger #10306: Issue comment #3135 (comment) created by greptile-apps Bot
7s
7s
[PyTorch][torch.compile] Decouple amax reduction group from the quantizer TE-CI Trigger #10305: Issue comment #3104 (comment) created by pggPL
3m 43s
3m 43s
Single-launch CUTLASS grouped GEMM for per-tensor NVFP4 TE-CI Trigger #10304: Issue comment #3134 (comment) created by greptile-apps Bot
1s
1s
Abstract CUDA hardcodes into configurable te_device_type / te_platform TE-CI Trigger #10303: Issue comment #3113 (comment) created by lxd-cumt
1s
1s
Abstract CUDA hardcodes into configurable te_device_type / te_platform TE-CI Trigger #10302: Issue comment #3113 (comment) created by lxd-cumt
2s
2s
NVFP4: cache GEMM-swizzled weight scale factors across micro-batches TE-CI Trigger #10301: Issue comment #3093 (comment) created by cael-ling
2s
2s
Enable NVFP4 RHT amax for grouped SReLU MLP TE-CI Trigger #10300: Issue comment #3133 (comment) created by greptile-apps Bot
3s
3s
[PyTorch] Reduce CPU overhead in grouped MLP block TE-CI Trigger #10299: Issue comment #2897 (comment) created by zhongbozhu
2s
2s
[Pytorch][Common] Hybrid quantization TE-CI Trigger #10298: Issue comment #2817 (comment) created by negvet
3m 49s
3m 49s
[torch.compile] Bunch of small changes needed for enabling torch.compile TE-CI Trigger #10297: Issue comment #3130 (comment) created by pggPL
3m 50s
3m 50s
Make NVTE tensor handle pool size configurable TE-CI Trigger #10296: Issue comment #3090 (comment) created by lhb8125
8s
8s
Introduce Mega-C++ to reduce CPU overhead TE-CI Trigger #10295: Issue comment #3099 (comment) created by zhongbozhu
3m 46s
3m 46s
[Common] Support scaled & clamped swiglu, srelu for BF16 TE-CI Trigger #10294: Issue comment #3132 (comment) created by zhongbozhu
3m 53s
3m 53s
[PyTorch] Update cuBLASLt grouped gemm filter TE-CI Trigger #10293: Issue comment #3119 (comment) created by yaox12
2s
2s
Introduce Mega-C++ to reduce CPU overhead TE-CI Trigger #10292: Issue comment #3099 (comment) created by greptile-apps Bot
2s
2s
[Common] Support scaled & clamped swiglu, srelu for BF16 TE-CI Trigger #10291: Issue comment #3132 (comment) created by greptile-apps Bot
2s
2s
Introduce Mega-C++ to reduce CPU overhead TE-CI Trigger #10290: Issue comment #3099 (comment) created by zhongbozhu
1s
1s
[PyTorch] Update cuBLASLt grouped gemm filter TE-CI Trigger #10289: Issue comment #3119 (comment) created by vthumbe1503
1s
1s
[PyTorch] Update cuBLASLt grouped gemm filter TE-CI Trigger #10288: Issue comment #3119 (comment) created by vthumbe1503
3m 51s
3m 51s
Current Scaling Group Quantization + Enabling Varying Last/Both Dims in Group Quantize TE-CI Trigger #10287: Issue comment #3114 (comment) created by vthumbe1503
3m 44s
3m 44s
[JAX] Remove shard_map from MoEBlock to support quant before FSDP AG using Grouped quant+GEMM custom partitioning rules TE-CI Trigger #10286: Issue comment #3131 (comment) created by greptile-apps Bot
2s
2s
[fix] Respect per-op activation-offload markers in fused grouped MLP TE-CI Trigger #10285: Issue comment #3128 (comment) created by fanshiqing
1s
1s