This is an automated email from the ASF dual-hosted git repository.
github-bot pushed a change to branch nightly
in repository https://gitbox.apache.org/repos/asf/tvm.git
from 8b4df725b7 [Runtime][KVCache] Initial interface setup for MLA (#17616)
add d93f4ad4b8 [KVCache] Add KV Cache for CPU Runtime (#17615)
No new revisions were added by this update.
Summary of changes:
python/tvm/relax/frontend/nn/llm/kv_cache.py | 882 ++++++++++++++++++++-
python/tvm/relax/frontend/nn/llm/tree_attn.py | 405 ++++++++++
src/runtime/cpu_device_api.cc | 48 ++
src/runtime/relax_vm/paged_kv_cache.cc | 3 +-
...untime_builtin_paged_attention_kv_cache_cpu.py} | 77 +-
5 files changed, 1353 insertions(+), 62 deletions(-)
copy tests/python/relax/{test_runtime_builtin_paged_attention_kv_cache_tir.py
=> test_runtime_builtin_paged_attention_kv_cache_cpu.py} (95%)
