Skip to content

Pull requests: alibaba/rtp-llm

New pull request New
59 Open 274 Closed
59 Open 274 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: support trtllm_fp4_block_scale_routed_moe
#446 opened Dec 11, 2025 by CrimsonDump Loading…
Feature/fix flexlb rolling issue
#445 opened Dec 11, 2025 by sunmiaozju Loading…
1
feat: support python-xqa with CUDA 12.9 and compatible with CUDA 12.6
#444 opened Dec 11, 2025 by qqbbiu Loading…
3
hotfix: fix try catch
#442 opened Dec 10, 2025 by wanglining97 Loading…
hotfix: reuse cache hit bug
#441 opened Dec 10, 2025 by MMadhatter Loading…
refactor(rtp_llm): fix frontend loop stuck issue
#438 opened Dec 10, 2025 by sunmiaozju Loading…
5
fix - fix auto config deepep
#437 opened Dec 9, 2025 by alibaba-miji Loading…
feat: add process-isolated logging with rank_id and server_id
#436 opened Dec 5, 2025 by sunmiaozju Loading…
fix: modify pre_decoder_residual under multimodalEmbedding input
#433 opened Dec 5, 2025 by junna2016 Loading…
17
[draft] master 0.0.2
#427 opened Dec 4, 2025 by jianglan89 Loading…
2
fix: wrong residual when pre decoder layernorm + mm embedding + quant
#426 opened Dec 4, 2025 by LLLLKKKK Loading…
1
feat: support pure tp + ep for cuda graph
#425 opened Dec 3, 2025 by JackTan25 Loading…
13
fix - backend server not shutdown graceful in mulit rank case
#424 opened Dec 3, 2025 by jianglan89 Loading…
5
support fp8 fmha for rocm pymodel
#422 opened Dec 3, 2025 by liaocz Loading…
1
feat: support return raw output and output ids in debug info
#421 opened Dec 2, 2025 by soaringk Loading…
5
refactor: optimize token reorder impl
#419 opened Dec 2, 2025 by MMadhatter Loading…
feature - add viztracer for inference api
#417 opened Dec 2, 2025 by jianglan89 Loading…
3
Support DeepSeek v3.2 encoding module
#415 opened Dec 2, 2025 by soaringk Loading…
8
refactor gemm
#414 opened Dec 1, 2025 by fff-2013 Loading…
2
fix: fix max context batch size
#412 opened Nov 28, 2025 by JackTan25 Loading…
Performance Optimization for Beam Search
#411 opened Nov 28, 2025 by zhangjianning-zjn Loading…
6
fix: hold host buffer util next forward
#410 opened Nov 28, 2025 by Vinkle-hzt Loading…
2
feat: handle cuda oom error for py model
#406 opened Nov 26, 2025 by MMadhatter Loading…
fix: fix custom_ar bug for rocm
#402 opened Nov 25, 2025 by liaocz Loading…
8
feat: update ar & fuse mla reuse cache
#399 opened Nov 25, 2025 by Nancheng-11 Loading…
5
ProTip! Filter pull requests by the default branch with base:main.