Repository navigation

Fix GQA shape inference for shape-derived int32 sequence lengths - #33884

Open

Tianlei Wu (tianleiwu) with Copilot wants to merge 8 commits into

mainfrom

copilot/fix-group-query-attention-load

Open

Fix GQA shape inference for shape-derived int32 sequence lengths#33884
Tianlei Wu (tianleiwu) with Copilot wants to merge 8 commits into
mainfrom
copilot/fix-group-query-attention-load

Commits on Nov 4, 2026
- Initial plan
  Copilot
  authored
  5de296e
- Add GQA shape-derived int32 sequence length regression
  
  Copilot
  and
  Tianlei Wu (tianleiwu)
  authored
  af69459
- Fix GQA shape inference for propagated int32 sequence lengths
  
  Copilot
  and
  Tianlei Wu (tianleiwu)
  authored
  9d33ad9
Commits on Nov 9, 2026
- Merge origin/main to refresh GQA shape-inference CI
  
  Tianlei Wu (tianleiwu)
  and
  Copilot
  committed
  c38ef55
- test(react-native): tap EPContext button by label
  
  Tianlei Wu (tianleiwu)
  and
  Copilot
  committed
  3fba4cf
- fix(react-native): finish worker before dispatch
  
  Tianlei Wu (tianleiwu)
  and
  Copilot
  committed
  e27bfcf
- fix(react-native): use unique Detox button ID
  
  Tianlei Wu (tianleiwu)
  and
  Copilot
  committed
  f3b4560
- ci(react-native): prepare selected simulator
  
  Tianlei Wu (tianleiwu)
  and
  Copilot
  committed
  85f9bf1