This website requires JavaScript.
Explore
Help
Sign In
iHore
/
vLLM
Watch
8
Star
0
Fork
0
Code
Issues
Pull Requests
Actions
1
Packages
Projects
Releases
Wiki
Activity
Files
235366fe2eb3144321978e181af94487f0215595
vLLM
/
vllm
/
engine
/
output_processor
T
History
youkaichao
and
Zhuohan Li
4fdc581f9e
[core] simplify seq group code (
#9569
)
...
Co-authored-by: Zhuohan Li <
zhuohan123@gmail.com
>
2024-10-24 00:16:44 -07:00
..
__init__.py
[Speculative decoding 6/9] Integrate speculative decoding with LLMEngine (
#3894
)
2024-04-16 13:09:21 -07:00
interfaces.py
[Bugfix] Fix incorrect updates to num_computed_tokens in multi-step scheduling (
#9038
)
2024-10-06 12:48:11 -07:00
multi_step.py
[CI/Build] mypy: Resolve some errors from checking vllm/engine (
#9267
)
2024-10-16 22:55:59 +00:00
single_step.py
[core] simplify seq group code (
#9569
)
2024-10-24 00:16:44 -07:00
stop_checker.py
[CI/Build] mypy: Resolve some errors from checking vllm/engine (
#9267
)
2024-10-16 22:55:59 +00:00
util.py
[CI/Build] mypy: Resolve some errors from checking vllm/engine (
#9267
)
2024-10-16 22:55:59 +00:00