Commits · 2cc206997012057152f194c0f25e19e3ab3297ea · OpenDAS / vllm_cscc · GitLab

25 Jun, 2025 22 commits
- [TPU][Bugfix] fix kv cache padding (#20048) · 2cc20699
  Chengji Yao authored Jun 25, 2025
```
Signed-off-by: Chengji Yao <chengjiyao@google.com>
```
  2cc20699
- [Bugfix] default set cuda_graph_sizes to max_num_seqs for v1 engine (#20062) · 9f0608fc
  zhrrr authored Jun 26, 2025
```
Signed-off-by: izhuhaoran <izhuhaoran@qq.com>
```
  9f0608fc
- Fix the path to the testing script. (#20082) · 4e0db57f
  QiliangCui authored Jun 25, 2025
```
Signed-off-by: Qiliang Cui <derrhein@gmail.com>
```
  4e0db57f
- [Misc] Add parallel state `node_count` function (#20045) · c40692bf
  Nick Hill authored Jun 25, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  c40692bf
- [PD] let toy proxy handle /chat/completions (#19730) · 4734704b
  lkchen authored Jun 25, 2025
```
Signed-off-by: Linkun <github@lkchen.net>
```
  4734704b
- static_scaled_fp8_quant should not run when scale.numel is not 1 (#20076) · 8b8c209e
  Eldar Kurtić authored Jun 25, 2025
  
  8b8c209e
- [Fix] Support cls pooling in ModernBertPooler (#20067) · 23a04e08
  lsz05 authored Jun 26, 2025
```
Signed-off-by: shengzhe.li <shengzhe.li@sbintuitions.co.jp>
```
  23a04e08
- [Quantization] Add compressed-tensors emulations support for NVFP4 (#19879) · 02c97d9a
  Dipika Sikka authored Jun 25, 2025
```
Signed-off-by: Dipika Sikka <dipikasikka1@gmail.com>
Signed-off-by: Dipika <dipikasikka1@gmail.com>
```
  02c97d9a
- [Frontend] Add `/v1/audio/translations` OpenAI API endpoint (#19615) · e795d723
  Nicolò Lucchesi authored Jun 25, 2025
```
Signed-off-by: Roger Wang <ywang@roblox.com>
Signed-off-by: NickLucche <nlucches@redhat.com>
Co-authored-by: Roger Wang <ywang@roblox.com>
```
  e795d723
- [V1][Speculative Decoding] Fix DeepSeek MTP (#20022) · 8359f4c8
  cjackal authored Jun 26, 2025
```
Signed-off-by: cjackal <44624812+cjackal@users.noreply.github.com>
```
  8359f4c8
- [Doc] Guide for Incremental Compilation Workflow (#19109) · bf518158
  Michael Goin authored Jun 25, 2025
  
  bf518158
- [doc] add reference link for Intel XPU (#20064) · c53fec1f
  Reid authored Jun 25, 2025
```
Signed-off-by: reidliu41 <reid201711@gmail.com>
Co-authored-by: reidliu41 <reid201711@gmail.com>
```
  c53fec1f
- [BugFix] Fix full-cuda-graph illegal memory access in FA3 (#20057) · 0f9e7354
  Lucas Wilkinson authored Jun 25, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  0f9e7354
- [Chore] debloat some initial logs (#19438) · ba7ba35c
  Aaron Pham authored Jun 25, 2025
```
Signed-off-by: Aaron Pham <contact@aarnphm.xyz>
```
  ba7ba35c
- [Kernels][Bugfix] Use torch op for all kernels in FusedMoE forward. Add... · 015fab8c
  bnellnm authored Jun 25, 2025
```
[Kernels][Bugfix] Use torch op for all kernels in FusedMoE forward.  Add additional testing for cudagraphs. (#19717)
Signed-off-by: Bill Nell <bnell@redhat.com>
```
  015fab8c
- [Feat][CLI] enforce-include-usage (#19695) · f59fc60f
  Max Wittig authored Jun 25, 2025
```
Signed-off-by: Max Wittig <max.wittig@siemens.com>
```
  f59fc60f
- [Refactor] Remove duplicate `ceil_div` (#20023) · 879f69be
  Wentao Ye authored Jun 25, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  879f69be
- [Frontend] speed up import time of vllm.config (#18036) · 71089341
  David Xia authored Jun 25, 2025
```
Signed-off-by: David Xia <david@davidxia.com>
```
  71089341
- Move to a faster base64 implementation (#19984) · 3443aaf8
  h-avsha authored Jun 25, 2025
```
Signed-off-by: h-avsha <avshalom.manevich@hcompany.ai>
```
  3443aaf8
- Revert "Fix(models/siglip): Add compatibility for Gemma models quantized by... · 2273ec32
  Isotr0py authored Jun 25, 2025
```
Revert "Fix(models/siglip): Add compatibility for Gemma models quantized by llm-compressor" (#20030)
```
  2273ec32
- Revert "[Feature] Integrate new deepgemm (#19820)" (#20049) · a6c4b87f
  Wentao Ye authored Jun 24, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  a6c4b87f
- [Llama4] Update `attn_temperature_tuning` (#19997) · 1afa9948
  Brayden Zhong authored Jun 24, 2025
```
Signed-off-by: Brayden Zhong <b8zhong@uwaterloo.ca>
```
  1afa9948
24 Jun, 2025 16 commits
- cmake: Update vllm_flash_attn for vllm_kernels (#20032) · 0d06b533
  Eli Uriegas authored Jun 24, 2025
```
Signed-off-by: Eli Uriegas <eliuriegas@meta.com>
```
  0d06b533
- use .dev for version comparison with pytorch nightly release (#20031) · c01d1c5a
  Boyuan Feng authored Jun 24, 2025
```
Signed-off-by: Boyuan Feng <boyuan@meta.com>
```
  c01d1c5a
- [Easy] Remove submodule added in #19463 (#20039) · ead36984
  Brayden Zhong authored Jun 24, 2025
```
Signed-off-by: Brayden Zhong <b8zhong@uwaterloo.ca>
```
  ead36984
- [Feature] Integrate new deepgemm (#19820) · c6e3bba8
  Wentao Ye authored Jun 24, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  c6e3bba8
- [P/D] Asynchronously do _nixl_handshake (#19836) · 91f7d9d0
  lkchen authored Jun 24, 2025
```
Signed-off-by: Linkun Chen <github@lkchen.net>
Signed-off-by: Nick Hill <nhill@redhat.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
```
  91f7d9d0
- [BugFix] Fix multi-node offline data parallel (#19937) · 8619e715
  Nick Hill authored Jun 24, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  8619e715
- [Misc][Benchmarking] Add variable request-rate ("ramp-up") to the benchmarking client. (#19423) · c635c5f7
  d.transposed authored Jun 24, 2025
```
Signed-off-by: dtransposed <damian@damian-ml-machine.europe-west3-b.c.jetbrains-grazie.internal>
Co-authored-by: dtransposed <damian@damian-ml-machine.europe-west3-b.c.jetbrains-grazie.internal>
Co-authored-by: Roger Wang <hey@rogerw.me>
```
  c635c5f7
- [Perf] Improve/Fix-regression for FA3 in High QPS regimes (#19463) · a045b7e8
  Lucas Wilkinson authored Jun 24, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
```
  a045b7e8
- [Fix][V1] Remove --scheduling-policy oracle (#20010) · 981eeca4
  amit authored Jun 24, 2025
```
Signed-off-by: amit <amit.man@gmail.com>
```
  981eeca4
- refactor example - qwen3_reranker (#19847) · 26d34eb6
  Reid authored Jun 24, 2025
```
Signed-off-by: reidliu41 <reid201711@gmail.com>
Co-authored-by: reidliu41 <reid201711@gmail.com>
```
  26d34eb6
- [Bugfix][CPU] Fix InputBatch for pooling models in the CPU v1 (#20014) · 53da4cd3
  Li, Jiang authored Jun 24, 2025
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
  53da4cd3
- [PERF] Speedup of MRoPE prepare inputs (#19939) · 9a3b8832
  Vadim Gimpelson authored Jun 24, 2025
```
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@centml.ai>
```
  9a3b8832
- add some examples for other benchmark scripts (#19893) · 3014c920
  Reid authored Jun 24, 2025
```
Signed-off-by: reidliu41 <reid201711@gmail.com>
Co-authored-by: reidliu41 <reid201711@gmail.com>
```
  3014c920
- [doc] Fix broken link in the installation for CPU (#19980) · 0eed5169
  Kay Yan authored Jun 24, 2025
```
Signed-off-by: Kay Yan <kay.yan@daocloud.io>
```
  0eed5169
- [Misc][Tools][Benchmark] Add profile to autotune script (#19711) · ee5ad8d2
  Chenyaaang authored Jun 23, 2025
```
Signed-off-by: Chenyaaang <chenyangli@google.com>
```
  ee5ad8d2
- Update test case parameter to have the throughput above 8.0 (#19994) · a738dbb2
  QiliangCui authored Jun 23, 2025
```
Signed-off-by: Qiliang Cui <derrhein@gmail.com>
```
  a738dbb2
23 Jun, 2025 2 commits
- [TPU] Fix tpu model runner test (#19995) · 33d5e29b
  Chenyaaang authored Jun 23, 2025
```
Signed-off-by: Chenyaaang <chenyangli@google.com>
```
  33d5e29b
- [Bugfix][Benchmark] Fix Marlin benchmark (#19929) · 4671ac6e
  22quinn authored Jun 23, 2025
  
  4671ac6e