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
e773247f8aee272d0262928117e3049cd2902b5e
vLLM
/
vllm
/
model_executor
/
model_loader
/
weight_cache
T
History
Xun Sun
e773247f8a
[Fast Start] Wait for weight cache daemon readiness (
#58370
)
2026-09-24 19:18:04 +00:00
..
__init__.py
[Fast Start] Cache the MTP draft model in a separate daemon group (
#57312
)
2026-09-22 03:13:38 +00:00
daemon.py
[Fast Start] Support data parallelism in the weight cache daemon (
#57386
)
2026-09-22 18:19:40 +00:00
ipc_loader.py
[Fast Start] Wait for weight cache daemon readiness (
#58370
)
2026-09-24 19:18:04 +00:00
protocol.py
[Fast Start] Support data parallelism in the weight cache daemon (
#57386
)
2026-09-22 18:19:40 +00:00
utils.py
[Fast Start] Cache the MTP draft model in a separate daemon group (
#57312
)
2026-09-22 03:13:38 +00:00