All workflows

Showing runs from all workflows

2500+ workflow runs

feat(wan): fast serving with persistent AOT caching and tuned inference recipe
completed successfully:
GitHub Actions Scan#879:Pull request#479synchronizebyPerseus14
Oct 1, 202613s
feat(attention): fixed-m splash attention kernel with dynamic bounds and safety fallbacks
completed successfully:
GitHub Actions Scan#878:Pull request#477synchronizebyPerseus14
Oct 1, 202616s
perf(wan): pad sequence to 128-lane multiple for self-attention (wan_seq_pad="lane")
completed successfully:
GitHub Actions Scan#877:Pull request#491synchronizebyPerseus14
Oct 1, 202612s
feat(attention): 2D Ulysses+Ring custom attention with fixed-m (pre-a2a norm reduction, unpadded ring K/V)
completed successfully:
GitHub Actions Scan#876:Pull request#478synchronizebyPerseus14
Oct 1, 202614s
feat(wan): fast serving with persistent AOT caching and tuned inference recipe
completed successfully:
GitHub Actions Scan#875:Pull request#479synchronizebyPerseus14
Oct 1, 202617s
feat(wan): fused RMSNorm+RoPE Pallas producer; Wan switches in config, not env
completed successfully:
GitHub Actions Scan#874:Pull request#488synchronizebyPerseus14
Oct 1, 202615s
feat(wan): fast inference optimizations (fused cross-attention, token…
completed successfully:
Linter#4896:Commit08ccfb2pushedbyPerseus14
feat/wan-fast-inference-optimizations
Oct 1, 20261m 37s
feat(attention): fixed-m splash attention kernel with dynamic bounds …
completed successfully:
Linter#4895:Commit1069267pushedbyPerseus14
feat/fixed-m-kernel
Oct 1, 20261m 24s
feat(attention): 2D Ulysses+Ring custom attention with fixed-m (pre-a…
completed successfully:
Linter#4894:Commita37cda4pushedbyPerseus14
feat/ring-attention
Oct 1, 20261m 36s
feat(wan): fused RMSNorm+RoPE Pallas producer; Wan switches in config…
completed successfully:
Linter#4893:Commitae05c1apushedbyPerseus14
feat/wan-custom-kernels
Oct 1, 20261m 24s
feat(wan): fast serving with persistent AOT caching and tuned inferen…
completed successfully:
Linter#4892:Commite6385e2pushedbyPerseus14
feat/wan-fast-serving
Oct 1, 20262m 49s
Add Pull Ready Label
completed successfully:
Add Pull Ready Label#7395:completedbyjitendra-jalwaniya
Oct 1, 20267s
Add Pull Ready Label
completed successfully:
Add Pull Ready Label#7394:completedbyjitendra-jalwaniya
Oct 1, 202611s
[PR 2/2] LTX2: SVG config and pipeline
completed successfully:
Unit Test#6099:Pull request#498synchronizebyjitendra-jalwaniya
ltx2_svg_pipeline
Oct 1, 20261h 28m 7s
ltx2: wire SVG config through pipeline, configs, AOT metadata, docs
completed successfully:
Linter#4891:Commit1b937b7pushedbyjitendra-jalwaniya
ltx2_svg_pipeline
Oct 1, 20261m 22s
[PR 1/2] SVG implementation for LTX 2
completed successfully:
Unit Test#6098:Pull request#497synchronizebyjitendra-jalwaniya
ltx2_svg_model
Oct 1, 20262h 51m 59s
[PR 1/2] SVG implementation for LTX 2
completed successfully:
Google GitHub Admin: Actions Workflow Security Scan#1529:Pull request#497synchronizebyjitendra-jalwaniya
ltx2_svg_model
Oct 1, 202647s
[PR 1/2] SVG implementation for LTX 2
completed successfully:
GitHub Actions Scan#873:Pull request#497synchronizebyjitendra-jalwaniya
Oct 1, 202613s
ltx2/svg: address #497 review (shared SVG helpers, config cleanup, te…
completed successfully:
Linter#4890:Commit2a40eb8pushedbyjitendra-jalwaniya
ltx2_svg_model
Oct 1, 20264m 18s
perf(wan2.2): Optimize attention kernels, rematerialization, and Ulysses parallelism for WAN 2.2 Training on TPU 7x
skipped:
Gemini Dispatch#1825:Pull request#487labeledbyToshi-31
Toshi-31:wan-2.2-perf-optimizations
Oct 1, 20261s
Add Pull Ready Label
completed successfully:
Add Pull Ready Label#7393:completedbyToshi-31
Oct 1, 20267s
Add Pull Ready Label
completed successfully:
Add Pull Ready Label#7392:completedbyPerseus14
Oct 1, 20269s
Add Pull Ready Label
completed successfully:
Add Pull Ready Label#7391:completedbyPerseus14
Oct 1, 20268s
Add Pull Ready Label
completed successfully:
Add Pull Ready Label#7390:completedbyPerseus14
Oct 1, 20269s
Add Pull Ready Label
completed successfully:
Add Pull Ready Label#7389:completedbyPerseus14
Oct 1, 20266s

Navigation Menu

All workflows

2500+ workflow runs

feat(wan): fast serving with persistent AOT caching and tuned inference recipe

feat(attention): fixed-m splash attention kernel with dynamic bounds and safety fallbacks

perf(wan): pad sequence to 128-lane multiple for self-attention (wan_seq_pad="lane")

feat(attention): 2D Ulysses+Ring custom attention with fixed-m (pre-a2a norm reduction, unpadded ring K/V)

feat(wan): fast serving with persistent AOT caching and tuned inference recipe

feat(wan): fused RMSNorm+RoPE Pallas producer; Wan switches in config, not env

feat(wan): fast inference optimizations (fused cross-attention, token…

feat(attention): fixed-m splash attention kernel with dynamic bounds …

feat(attention): 2D Ulysses+Ring custom attention with fixed-m (pre-a…

feat(wan): fused RMSNorm+RoPE Pallas producer; Wan switches in config…

feat(wan): fast serving with persistent AOT caching and tuned inferen…

Add Pull Ready Label

Add Pull Ready Label

[PR 2/2] LTX2: SVG config and pipeline

ltx2: wire SVG config through pipeline, configs, AOT metadata, docs

[PR 1/2] SVG implementation for LTX 2

[PR 1/2] SVG implementation for LTX 2

[PR 1/2] SVG implementation for LTX 2

ltx2/svg: address #497 review (shared SVG helpers, config cleanup, te…

perf(wan2.2): Optimize attention kernels, rematerialization, and Ulysses parallelism for WAN 2.2 Training on TPU 7x

Add Pull Ready Label

Add Pull Ready Label

Add Pull Ready Label

Add Pull Ready Label

Add Pull Ready Label

ProTip! You can narrow down the results and go further in time using the filter or the other filters available.

Workflow runs

2500+ workflow runs

ProTip! You can narrow down the results and go further in time using the created: filter or the other filters available.

ProTip! You can narrow down the results and go further in time using the filter or the other filters available.