Skip to content

Pull requests: taco-project/FlexKV

New pull request New
22 Open 189 Closed
22 Open 189 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
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Sglang + flexkv to run glm5.2 with nvfp4 kvcache type
#226 opened Jul 16, 2026 by linhu-nv Collaborator Loading…
feat(dsv4): add DeepSeek-V4 KV cache and layerwise transfer support
#225 opened Jul 15, 2026 by XingLiu1 Collaborator Loading…
22
Moritzxu/swa refactor
#224 opened Jul 15, 2026 by Luis-xu Collaborator • Draft
1
main mooncakestore merge
#218 opened Jul 13, 2026 by LfieLike Loading…
Add Mooncake Store backend support and SGLang compatibility aliases
#217 opened Jul 13, 2026 by LfieLike Loading…
Add and harden FlexKV stress benchmark
#214 opened Jul 12, 2026 by XingLiu1 Collaborator • Draft
use single concrete FlexKVConnectorStats subclass for worker and scheduler
#189 opened Jun 26, 2026 by linhu-nv Collaborator Loading…
integrating mooncake store backend for kv cache sharing.
#181 opened Jun 16, 2026 by Luis-xu Collaborator Loading…
fix: correct per-PP-stage CPU/SSD layer offset and add transfer correctness test suite
#171 opened May 21, 2026 by zhjc1124 Contributor • Draft
feat: unify parallel dimension abstraction and support sglang DP Attention
#170 opened May 20, 2026 by zhjc1124 Contributor • Draft
feat(dist_reuse): KV cache sharing across TP/PP/CP + single-node radi…
#169 opened May 19, 2026 by feiqiangs Collaborator Loading…
14
Prefetch optimize, adding three prefetch strategies
#163 opened May 12, 2026 by Luis-xu Collaborator Loading…
use wider copy granuality and bypass L1
#157 opened Apr 29, 2026 by linhu-nv Collaborator Loading…
Gemma4 support
#152 opened Apr 18, 2026 by linhu-nv Collaborator Loading…
flexkv server client opt draft
#144 opened Apr 13, 2026 by linhu-nv Collaborator Loading…
[Feat] nvcomp integration
#140 opened Apr 7, 2026 by zhuofan1123 Collaborator • Draft
[feature] Add SGLang HiCacheStorage integration with local and distributed modes
#137 opened Apr 5, 2026 by staryxchen Collaborator Loading…
7 tasks done
feat: add multi-GPU backend abstraction layer (gpu_backend)
#131 opened Mar 31, 2026 by feiqiangs Collaborator Loading…
feat: add MUSA support for FlexKV
#126 opened Mar 23, 2026 by superleo Loading…
1
simm integration into flexkv
#115 opened Mar 4, 2026 by linhu-nv Collaborator • Draft
2
add bypass cpu capability for local transfer
#104 opened Feb 9, 2026 by didoteebin Loading…
1
Feat/layerwise
#91 opened Jan 12, 2026 by zhuofan1123 Collaborator • Draft
ProTip! Follow long discussions with comments:>50.