Skip to content

Issues: kvcache-ai/Mooncake

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

18 Open 28 Closed
18 Open 28 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

p2p store transmission time?
#81 opened Jan 14, 2025 by skyCreateXian
proposal: improve the topology metadata
#78 opened Jan 13, 2025 by doujiang24
2
Libfabric transport layer support
#76 opened Jan 10, 2025 by xendo
1
Questions about transfer_engine_bench
#75 opened Jan 10, 2025 by cyLi-Tiger
1
proposal: topology discover & install transport automatically
#69 opened Jan 8, 2025 by doujiang24
1
How to set protocol parameter when prefill and decode in the same node?
#68 opened Jan 7, 2025 by gujingit
2
Why do we need to disable CUDA graph (enforce_eager) for Tensor parallelism?
#64 opened Jan 6, 2025 by skyCreateXian
1
Why is ITL's first token so long?
#62 opened Jan 3, 2025 by sunshenao
3
对于p2p-store-example，用阿里云eRDMA，设置MC_GID_INDEX=1时无法启动，只有0才能启动，不过ibv_modify_qp仍然报错
#53 opened Dec 26, 2024 by power-more
1
Inquiry Regarding "Mooncake- A KVCache-centric Disaggregated Architecture for LLM Serving"
#50 opened Dec 25, 2024 by VegetaPn
16
CXL/Shared memory
#48 opened Dec 22, 2024 by stmatengss
@stmatengss
2
[RoadMap] Mooncake Roadmap Q1 & Q2 2025 Roadmap Future roadmap or plan for new features
#44 opened Dec 18, 2024 by stmatengss
2 of 39 tasks
@james0zan
8
"Address not registered by any device(s)" Error when block_size is large
#38 opened Dec 16, 2024 by c-guo16
2
vllm-integration with multi rdma devices error
#35 opened Dec 13, 2024 by junna2016
17
Mooncake Transfer Engine 支持华为的Ascend昇腾平台和NPU显卡吗
#34 opened Dec 13, 2024 by zhaofangtao
1
Remote prefilling
#24 opened Dec 9, 2024 by toilaluan
1
benchmark performance
#5 opened Nov 28, 2024 by BaiStone2017
3
Inquiry Regarding Prefix Caching in the vllm Baseline
#2 opened Jul 27, 2024 by Lin-Qingyang-Alec
1
ProTip! Adding no:label will show everything without a label.