Skip to content

Pull requests: alibaba/rtp-llm

New pull request New
127 Open 891 Closed
127 Open 891 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

Feat: Let RTP ROCm side use ATOM plugin interfaces
#1141 opened Jun 25, 2026 by zhaoan12-prc Collaborator Loading…
10
kvcache framework restructure
#1139 opened Jun 25, 2026 by Adrenaline-S Collaborator • Draft
feat(flexlb): batch scheduling infrastructure, strategy refactor, and error model hardening
#1138 opened Jun 24, 2026 by wzy-99 Collaborator Loading…
5
fix(bert): propagate multimodal features to Python BertModel forward
#1137 opened Jun 23, 2026 by parkerpang Collaborator Loading…
8
Feature/kvcm online optmizer
#1135 opened Jun 23, 2026 by YoungRX Collaborator Loading…
1
refactor(profiler): add RAII step scope for service-wide timeline
#1133 opened Jun 23, 2026 by psmarter Collaborator Loading…
2
fix: enable model analyzer to resolve DeepSeek-V4 and Qwen3.5 models
#1130 opened Jun 22, 2026 by kwohting Loading…
3
refactor(attn): merge host/device fields of PyAttentionInputs into Ho…
#1128 opened Jun 22, 2026 by Jonathan-hwx Loading…
10
feat: update flashinfer sampling
#1126 opened Jun 22, 2026 by Vinkle-hzt Collaborator Loading…
4
feat(deps): pip dep unification + cu126/dead-config cleanup + WORKSPA…
#1124 opened Jun 22, 2026 by baohengyi Loading…
10
feat: merge python_native_v2 into main with review fixes
#1121 opened Jun 21, 2026 by baohengyi Loading…
13
feat: add cpu tp broadcaster
#1119 opened Jun 18, 2026 by Vinkle-hzt Collaborator Loading…
12
Feat/model inputs dump
#1118 opened Jun 18, 2026 by JackTan25 Collaborator Loading…
9
fix: pin cache store workers to local device
#1116 opened Jun 17, 2026 by Vinkle-hzt Collaborator Loading…
9
feat(moe): lazy init + full-name fallback for DeepGEMM symbol resolution
#1115 opened Jun 17, 2026 by Honglei-Qiu Loading…
8
fix - [flexlb] queue time calc is wrong
#1114 opened Jun 17, 2026 by jianglan89 Collaborator Loading…
1
feat: jit cache management
#1112 opened Jun 17, 2026 by loomts Loading…
11
feat(moe): add bf16 DeepEP-normal MoE path via DeepGEMM grouped GEMM
#1111 opened Jun 17, 2026 by Tanmo-ai Loading…
11
feat(xpu): add Intel GPU (XPU) support
#1110 opened Jun 16, 2026 by aslanxie Loading…
57
Feature/python native v2
#1108 opened Jun 16, 2026 by baohengyi Loading…
15
feat: support async decode scheduling
#1107 opened Jun 16, 2026 by Vinkle-hzt Collaborator Loading…
3
wakeup output queue when stream is finished or errored
#1106 opened Jun 16, 2026 by yyhclimacool Contributor Loading…
4
feat: add RDMA transport for ViT
#1105 opened Jun 15, 2026 by ydshi0 Loading…
2
feat(cache): migrate DSV4 KV cache infrastructure from feat/dsv4_on_dev_refactor
#1103 opened Jun 15, 2026 by Adrenaline-S Collaborator • Draft
[ROCm] Prefill performance optimization for embedding models
#1102 opened Jun 15, 2026 by liaocz Collaborator Loading…
18
ProTip! Adding no:label will show everything without a label.