-
Notifications
You must be signed in to change notification settings - Fork 277
Pull requests: alibaba/rtp-llm
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
feat: allow online FP8 quantization with W4A16 dense FFN
#1454
opened Sep 22, 2026 by
HongminTan
Collaborator
Loading…
feat(rocm): single-launch verify attention, DFlash drafting, deterministic FP8 MoE and TorchSpec DSpark
#1453
opened Sep 22, 2026 by
yzyDavid
Collaborator
Loading…
fix(openai): release the parked think tail when generation stops inside a think block
#1451
opened Sep 21, 2026 by
Linxiushen
Loading…
feat: add Master HTTP batch fanout and batch placement
#1448
opened Sep 21, 2026 by
CHW0218
Collaborator
Loading…
feat: optimize dispatch, KV cache, RPC lifecycle, and ROCm sampling
#1447
opened Sep 21, 2026 by
zhangjianning-zjn
Collaborator
Loading…
feat(p2p): integrate planned decode handoff
#1443
opened Sep 18, 2026 by
ZhihanYan
Collaborator
Loading…
feat: migrate RTP-LLM to Python Native packaging and pytest execution
#1441
opened Sep 17, 2026 by
baohengyi
Loading…
feat(recording): record batch schedules and embedded request timing v…
#1440
opened Sep 17, 2026 by
HaiyuZhou
Loading…
fix(trace): prevent false FetchResponse cancellation after normal completion
#1438
opened Sep 17, 2026 by
SAzwj
Collaborator
Loading…
feat(multimodal): add standalone KVCM EMB Python client
#1436
opened Sep 16, 2026 by
charpty
Collaborator
Loading…
feat: Minimax-M3 model integration on main
#1433
opened Sep 16, 2026 by
qqbbiu
Collaborator
Loading…
feat: support context custom output processors
#1432
opened Sep 15, 2026 by
CHW0218
Collaborator
Loading…
perf: speed up KV cache block tracking
#1428
opened Sep 14, 2026 by
zhangjianning-zjn
Collaborator
Loading…
perf: speed up stop_word_id comparison
#1426
opened Sep 14, 2026 by
zhangjianning-zjn
Collaborator
Loading…
fix(think): align think handling with the template anchor and the seq budget
#1424
opened Sep 12, 2026 by
hy-guo
Loading…
refactor(cache): unify independent pools and group identity routing
#1423
opened Sep 12, 2026 by
Adrenaline-S
Collaborator
Loading…
feat(rocm): upgrade AITER and migrate kernels to Triton 3.8
#1421
opened Sep 11, 2026 by
junna2016
Collaborator
Loading…
fix(renderer): NameError in StreamStatusSync.update_output_sync — reference undefined
output
#1414
opened Sep 9, 2026 by
Anai-Guo
Loading…
fix(dsv4/fp8): import
suppress so FP8 indexer prefill cleanup stops masking the real error
#1412
opened Sep 9, 2026 by
Anai-Guo
Loading…
Previous Next
ProTip!
Follow long discussions with comments:>50.