Commit 68d80b1

committed

Add with_output version AppendAttention

1 parent c011cb8 commit 68d80b1Copy full SHA for 68d80b1

5 files changed

+707

-92

lines changed

custom_ops/gpu_ops
- append_attention.cu
- cpp_extensions.cc
fastdeploy/model_executor/layers/attention
- append_attn_backend.py
- ops
  - __init__.py
  - append_attention.py

5 files changed

+707

-92

lines changed

Comments

(0)