feat : add radix prefix cache
- replace hash-only lookup with page-granular radix matching - keep partial pages private and cache only materialized KV prefixes - integrate completed-request caching and add radix behavior tests
This commit is contained in:
@@ -5,7 +5,7 @@ from astrai.inference.core.cache import (
|
||||
KVCache,
|
||||
KVStorage,
|
||||
PagePool,
|
||||
PrefixCache,
|
||||
RadixCache,
|
||||
ReqToTokenPool,
|
||||
page_hash,
|
||||
)
|
||||
@@ -18,7 +18,7 @@ __all__ = [
|
||||
"KVCache",
|
||||
"KVStorage",
|
||||
"PagePool",
|
||||
"PrefixCache",
|
||||
"RadixCache",
|
||||
"ReqToTokenPool",
|
||||
"page_hash",
|
||||
"Executor",
|
||||
|
||||
Reference in New Issue
Block a user