Commits · 9dd945c1cb2f95df4e19b9e673e2be56cbe5caf5 · OpenDAS / vllm_cscc · GitLab

06 Aug, 2025 1 commit
- [feat]支持mtp模型full_cuda_graph · 9dd945c1
  王敏 authored Aug 06, 2025
  
  9dd945c1
05 Aug, 2025 2 commits
- [feat]优化mtp相关函数返回类型 · 7e71c143
  王敏 authored Aug 05, 2025
  
  7e71c143
- [feat]1.支持mtp模型 full_cuda_graph; 2.优化mtp拒绝采样 · 8e0ae19d
  王敏 authored Aug 05, 2025
  
  8e0ae19d
31 Jul, 2025 1 commit
- [feat]支持v1 engine mtp cudagraph · fe393be8
  王敏 authored Jul 31, 2025
  
  fe393be8
24 Jul, 2025 1 commit
- support full_cuda_graph · 8da1c576
  zhuwenwen authored Jul 24, 2025
  
  8da1c576
12 Jul, 2025 1 commit
- [fix]1.优化mtp代码，解决prefix-caching不兼容问题; 2.修复v1 engine从tokenizer config中读取max_model_len导致长文本输入报错 · b16f183c
  王敏 authored Jul 12, 2025
  
  b16f183c
07 Jul, 2025 1 commit
- [feat]支持v1 engine flashmla和mtp同时开启 · ac4dedb1
  王敏 authored Jul 07, 2025
  
  ac4dedb1
06 Jul, 2025 1 commit
- [BugFix][Spec Decode] Fix spec token ids in model runner (#20530) · 9528e3a0
  Woosuk Kwon authored Jul 06, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  9528e3a0
04 Jul, 2025 1 commit

Enable V1 for Hybrid SSM/Attention Models (#20016) · 2f35a022

Thomas Parnell authored Jul 04, 2025


Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>
Co-authored-by: Stanislaw Wozniak <stw@zurich.ibm.com>
Co-authored-by: Tyler Michael Smith <tysmith@redhat.com>
Co-authored-by: Chen Zhang <zhangch99@outlook.com>

2f35a022

02 Jul, 2025 1 commit

[V1] LogitsProcessor programming model (#16728) · 48fb076c

afeldman-nm authored Jul 02, 2025


Signed-off-by: Nick Hill <nhill@redhat.com>
Signed-off-by: Andrew Feldman <afeldman@neuralmagic.com>
Signed-off-by: Andrew Feldman <afeldman@redhat.com>
Co-authored-by: Nick Hill <nhill@redhat.com>

48fb076c

01 Jul, 2025 2 commits
- [Optimization] Cache sampled token ids in model runner (#20291) · 7f280d69
  Woosuk Kwon authored Jul 01, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  7f280d69
- [V1] Only print cudagraph tqdm on rank 0 with `is_global_first_rank` (#19516) · be250bbc
  Michael Goin authored Jul 01, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  be250bbc
30 Jun, 2025 3 commits
- [Optimization] Use Shared `CachedRequestData` Instance Across All Requests (#20232) · 2863befc
  Woosuk Kwon authored Jun 30, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  2863befc
- [Spec Decode] Refactor spec decoding into a separate function (#20238) · 2062c072
  Woosuk Kwon authored Jun 30, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  2062c072
- [Misc] Fix import (#20233) · 19108ef3
  Woosuk Kwon authored Jun 29, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  19108ef3
26 Jun, 2025 1 commit
- [Feature] Expert Parallelism Load Balancer (EPLB) (#18343) · e9fd658a
  Bowen Wang authored Jun 26, 2025
```
Signed-off-by: Bowen Wang <abmfy@icloud.com>
```
  e9fd658a
24 Jun, 2025 1 commit
- [PERF] Speedup of MRoPE prepare inputs (#19939) · 9a3b8832
  Vadim Gimpelson authored Jun 24, 2025
```
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@centml.ai>
```
  9a3b8832
23 Jun, 2025 1 commit
- [Bugfix][v1] Fix step pooler implementation and step pooling usage in v1 (#19956) · 61f4fc5d
  Isotr0py authored Jun 24, 2025
```
Signed-off-by: Isotr0py <2037008807@qq.com>
```
  61f4fc5d
20 Jun, 2025 1 commit
- Export NaNs in logits to scheduler_stats if output is corrupted (#18777) · 2e3e3c86
  Vlad Tiberiu Mihailescu authored Jun 20, 2025
```
Signed-off-by: Vlad Mihailescu <vtmihailescu@gmail.com>
```
  2e3e3c86
19 Jun, 2025 2 commits

Support embedding models in V1 (#16188) · 799397ee

Maximilien de Bayser authored Jun 19, 2025


Signed-off-by: Max de Bayser <mbayser@br.ibm.com>
Signed-off-by: Max de Bayser <maxdebayser@gmail.com>
Signed-off-by: 22quinn <33176974+22quinn@users.noreply.github.com>
Co-authored-by: 22quinn <33176974+22quinn@users.noreply.github.com>

799397ee

[BugFix] Fix use_cudagraph=False (#19612) · ed333497
Richard Zou authored Jun 18, 2025
```
Signed-off-by: Richard Zou <zou3519@gmail.com>
```
ed333497

18 Jun, 2025 1 commit
- [v1] Support mamba2 (#19327) · a89209b7
  Chen Zhang authored Jun 19, 2025
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  a89209b7
13 Jun, 2025 2 commits
- [CUDA] Enable full cudagraph for FlashMLA (#18581) · 3597b06a
  Luka Govedič authored Jun 13, 2025
```
Signed-off-by: luka <luka@neuralmagic.com>
```
  3597b06a
- [Fix] The zip function in Python 3.9 does not have the strict argument (#19549) · cefdb996
  汪志鹏 authored Jun 13, 2025
```
Signed-off-by: 汪志鹏 <wangzhipeng628@gmail.com>
```
  cefdb996
12 Jun, 2025 1 commit
- [V1] Resolve failed concurrent structured output requests (#19565) · c57bb199
  Russell Bryant authored Jun 12, 2025
```
Signed-off-by: Russell Bryant <rbryant@redhat.com>
```
  c57bb199
11 Jun, 2025 1 commit
- [UX] Add Feedback During CUDAGraph Capture (#19501) · 97a9465b
  Robert Shaw authored Jun 11, 2025
```
Signed-off-by: rshaw@neuralmagic.com <robertgshaw2@gmail.com>
```
  97a9465b
10 Jun, 2025 1 commit
- [Core] Batch multi modal input using pinned memory (#19169) · 319cb1e3
  Lukas Geiger authored Jun 10, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
  319cb1e3
09 Jun, 2025 2 commits
- [Misc] Fixes and Optimizations for DeepEP + DeepGEMM combination. (#19298) · 5cf2daea
  Varun Sundar Rabindranath authored Jun 09, 2025
```
Signed-off-by: Varun <vsundarr@redhat.com>
Co-authored-by: Varun <vsundarr@redhat.com>
```
  5cf2daea
- [full_graph] Fix query_start_loc padding (#19321) · 770e5dcd
  Yinghai Lu authored Jun 09, 2025
```
Signed-off-by: Yinghai Lu <yinghai@thinkingmachines.ai>
```
  770e5dcd
06 Jun, 2025 2 commits
- [BugFix] Fix tpu_model_runner block_id concatenation (#19228) · 46ecc579
  Nick Hill authored Jun 06, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  46ecc579
- [v1] Hybrid Memory Allocator (#17996) · f8a1a2d1
  Chen Zhang authored Jun 06, 2025
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  f8a1a2d1
05 Jun, 2025 1 commit
- [BugFix][Minor] Fix full cuda graph bug when max_num_seqs < 512 (#19171) · af7fc84f
  Woosuk Kwon authored Jun 04, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  af7fc84f
04 Jun, 2025 3 commits
- [Bugfix] Fix FA3 full cuda graph correctness (#19106) · b124e108
  Woosuk Kwon authored Jun 03, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  b124e108
- [Core] Cast multimodal input in hf processor (#18862) · 1409ef91
  Lukas Geiger authored Jun 04, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
  1409ef91
- [CPU] V1 support for the CPU backend (#16441) · 4555143e
  Li, Jiang authored Jun 04, 2025
  
  4555143e
03 Jun, 2025 3 commits
- [v1] Re-init input batch for multiple kv cache groups (#18654) · 6cac54f4
  Chen Zhang authored Jun 04, 2025
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  6cac54f4
- [V1] Support cross-layer KV sharing (#18212) · bdf13965
  Yong Hoon Shin authored Jun 03, 2025
```
Signed-off-by: Yong Hoon Shin <yhshin@meta.com>
```
  bdf13965
- [Misc] Add SPDX-FileCopyrightText (#19100) · 02f0c7b2
  Simon Mo authored Jun 03, 2025
```
Signed-off-by: simon-mo <simon.mo@hey.com>
```
  02f0c7b2
02 Jun, 2025 1 commit
- [Core] Support inplace model weights loading (#18745) · 9760fd8f
  22quinn authored Jun 02, 2025
```
Signed-off-by: 22quinn <33176974+22quinn@users.noreply.github.com>
```
  9760fd8f
01 Jun, 2025 1 commit

[Misc] reuse num_tokens_across_dp of get_dp_padding to avoid unnecessary dp... · d6fd3a33

zhrrr authored Jun 02, 2025


[Misc] reuse num_tokens_across_dp of get_dp_padding to avoid unnecessary dp all reduce in set_forward_context (#18935)
Signed-off-by: Tyler Michael Smith <tysmith@redhat.com>
Co-authored-by: zhuhaoran <zhuhaoran.zhr@alibaba-inc.com>
Co-authored-by: Tyler Michael Smith <tysmith@redhat.com>

d6fd3a33