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
e1bb2fd52dea0bbc772bdf35fd27664c5daec7b2
vLLM
/
vllm
/
core
/
block
T
History
Cade Daniel
e95cd87959
[Speculative decoding 6/9] Integrate speculative decoding with LLMEngine (
#3894
)
2024-04-16 13:09:21 -07:00
..
__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
[Speculative decoding 6/9] Integrate speculative decoding with LLMEngine (
#3894
)
2024-04-16 13:09:21 -07:00
common.py
[Core][Bugfix]Refactor block manager for better testability (
#3492
)
2024-03-27 23:59:28 -07:00
cpu_gpu_block_allocator.py
[Core][Bugfix]Refactor block manager for better testability (
#3492
)
2024-03-27 23:59:28 -07:00
interfaces.py
[Doc] Add typing hints / mypy types cleanup (
#3816
)
2024-04-11 17:17:21 -07:00
naive_block.py
[Core][Bugfix]Refactor block manager for better testability (
#3492
)
2024-03-27 23:59:28 -07:00
prefix_caching_block.py
[Core][Bugfix]Refactor block manager for better testability (
#3492
)
2024-03-27 23:59:28 -07:00