PyPI page
Home page
Author:
Mooncake Authors
Summary:
A KVCache-centric Disaggregated Architecture for large-scale LLM inference and training. (CUDA 13 version)
Latest version:
0.3.13.post1
Required dependencies:
aiohttp
|
msgpack
|
requests
|
typing-extensions
Downloads last day:
20,716
Downloads last week:
117,594
Downloads last month:
152,966
7-day smoothing