This website requires JavaScript.
Explore
Help
Sign In
iHore
/
vLLM
Watch
8
Star
0
Fork
0
Code
Issues
Pull Requests
Actions
2
Packages
Projects
Releases
Wiki
Activity
Files
e97f802b2d74861af77997691a7d1c36498f6dca
vLLM
/
vllm
/
core
/
block
T
History
Cody Yu
7206ce4ce1
[Core] Support
reset_prefix_cache
(
#12284
)
2025-01-22 18:52:27 +00:00
..
__init__.py
[Bugfix] Add
__init__.py
files for
vllm/core/block/
and
vllm/spec_decode/
(
#3798
)
2024-04-02 12:35:31 -07:00
block_table.py
fix: [doc] fix typo (
#11751
)
2025-01-06 00:48:16 +08:00
common.py
[Core] support LoRA and prompt adapter in content-based hashing for Block Manager v2 prefix caching (
#8240
)
2024-12-13 07:51:25 -08:00
cpu_gpu_block_allocator.py
[Core] Support
reset_prefix_cache
(
#12284
)
2025-01-22 18:52:27 +00:00
interfaces.py
[Core] Support
reset_prefix_cache
(
#12284
)
2025-01-22 18:52:27 +00:00
naive_block.py
[Core] Support
reset_prefix_cache
(
#12284
)
2025-01-22 18:52:27 +00:00
prefix_caching_block.py
[Core] Support
reset_prefix_cache
(
#12284
)
2025-01-22 18:52:27 +00:00
utils.py
[Core] Deprecating block manager v1 and make block manager v2 default (
#8704
)
2024-10-17 11:38:15 -05:00