Commit ecbc5fb

committed

Add with_output version AppendAttention

1 parent c011cb8 commit ecbc5fbCopy full SHA for ecbc5fb

7 files changed

+709

-94

lines changed

custom_ops/gpu_ops
- append_attention.cu
- cpp_extensions.cc
fastdeploy/model_executor/layers/attention
test/layers
- test_append_attention.py

7 files changed

+709

-94

lines changed

Comments

(0)