Skip to content

Pull requests: alibaba/rtp-llm

New pull request New
67 Open 382 Closed
67 Open 382 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

feature - use torch to check inf/nan and dump tensor when inf/nan
#562 opened Jan 13, 2026 by jianglan89 Loading…
5
support logit_bias generate config
#560 opened Jan 13, 2026 by zongyuanwu Loading…
deepgemm input fp8
#559 opened Jan 13, 2026 by netaddi Loading…
10
Amd/speculative sampling and mtp
#557 opened Jan 12, 2026 by amd-yilizhao Loading…
4
feat: support headwise attention
#556 opened Jan 12, 2026 by Echo-2334 Loading…
feat: add memory connector
#553 opened Jan 12, 2026 by li-xiao-qing Loading…
perf: use event to sync graph input instead of h2h copy
#551 opened Jan 12, 2026 by Vinkle-hzt Loading…
11
feat: support python-xqa with CUDA 12.9
#550 opened Jan 12, 2026 by qqbbiu Loading…
3
feat: Optimize PA for Low-Concurrency cases
#549 opened Jan 12, 2026 by yanglf1121 Loading…
1
fix: fix global info
#548 opened Jan 12, 2026 by wanglining97 Loading…
12
feat: support headwise attention
#545 opened Jan 9, 2026 by Echo-2334 Loading…
1
fix: fix frontend embedding endpoint deps
#544 opened Jan 8, 2026 by wanglining97 Loading…
5
feat: frontend generate unique req id
#541 opened Jan 8, 2026 by wanglining97 Loading…
31
feat: native setup.py and pytest
#537 opened Jan 7, 2026 by LLLLKKKK Loading…
Feat/sp prefill cuda graph
#536 opened Jan 7, 2026 by JackTan25 Loading…
5
feat: Support prompt list format and add flashinfer-python hash
#534 opened Jan 7, 2026 by Zoe923 Loading…
6
feat: convert all smoke to args
#529 opened Jan 6, 2026 by JackTan25 Loading…
2
fix - clear incomplete kv cache
#526 opened Jan 6, 2026 by jianglan89 Loading…
fix: add log (NOT MERGE)
#524 opened Jan 5, 2026 by JackTan25 Loading…
5
feat: support draft model vocab prune
#521 opened Jan 5, 2026 by Vinkle-hzt Loading…
1
Fix triton jit recompile for qwen3 next prefill
#519 opened Jan 4, 2026 by alibaba-miji Loading…
2
test: add mtp executor test
#517 opened Jan 4, 2026 by Vinkle-hzt Loading…
18
Feature/weight mask
#510 opened Dec 29, 2025 by zhanghuanzj Loading…
5
[wip] Feat/support prefill cp
#508 opened Dec 29, 2025 by MMadhatter Loading…
Feature/mtp reuse cache
#507 opened Dec 29, 2025 by xinfei-shi Loading…
6
ProTip! Mix and match filters to narrow down what you’re looking for.