onnxruntime
[CUDA] Default QMoE GEMV fp16 accumulation for fp16 activations
#29166
Merged

[CUDA] Default QMoE GEMV fp16 accumulation for fp16 activations #29166

tianleiwu
tianleiwu perf(cuda): default QMoE GEMV to fp16 accumulation
98ea71c7
tianleiwu tianleiwu changed the title Default QMoE GEMV fp16 accumulation for fp16 activations [CUDA] Default QMoE GEMV fp16 accumulation for fp16 activations 28 days ago
jiafatom jiafatom requested a review from copilot-pull-request-reviewer copilot-pull-request-reviewer 28 days ago
copilot-pull-request-reviewer
copilot-pull-request-reviewer commented on 2026-06-19
jiafatom
jiafatom approved these changes on 2026-06-19
tianleiwu tianleiwu enabled auto-merge (squash) 28 days ago
tianleiwu tianleiwu merged bceab03c into main 28 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone