Commits · ff08e51940a77d2dd14a6c512bec4613d060b4fa · OpenDAS / vllm_cscc · GitLab

30 Jul, 2025 18 commits
- [NVIDIA] Fix Llama4 Scout FP4 functionality issues (#21499) · ff08e519
  Po-Han Huang (NVIDIA) authored Jul 30, 2025
```
Signed-off-by: Po-Han Huang <pohanh@nvidia.com>
```
  ff08e519
- [Misc] Improve code readability of KVCacheManager (#21673) · 8f4a1c9a
  Ruixiang Tan authored Jul 30, 2025
```
Signed-off-by: tanruixiang <tanruixiang0104@gmail.com>
Signed-off-by: Ruixiang Tan <819464715@qq.com>
Signed-off-by: GitHub <noreply@github.com>
```
  8f4a1c9a
- [misc] skip p2p check by default (#21904) · e91d3c9c
  youkaichao authored Jul 30, 2025
  
  e91d3c9c
- [Feature] Support multiple api keys in server (#18548) · bf668b5b
  Yan Pashkovsky authored Jul 30, 2025
```
Signed-off-by: Yan Pashkovsky <yanp.bugz@gmail.com>
```
  bf668b5b
- [Bugfix] we should use metavar is not choices (#21902) · da3e0bd6
  rongfu.leng authored Jul 30, 2025
```
Signed-off-by: rongfu.leng <rongfu.leng@daocloud.io>
```
  da3e0bd6
- [Feature][EPLB] Add eplb support for Qwen3 (#20815) · d979dd6b
  aladerran authored Jul 30, 2025
```
Signed-off-by: aladerran <aladerran@gmail.com>
```
  d979dd6b
- [benchmark] add max-concurrency in result table (#21095) · 533db093
  Peter Pan authored Jul 30, 2025
```
Signed-off-by: Peter Pan <Peter.Pan@daocloud.io>
```
  533db093
- [Model] Remove DSV2 unused code (#21903) · fc91da54
  Jee Jee Li authored Jul 30, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  fc91da54
- [Misc] Remove redundant config definitions (#21891) · 2ca5f82c
  Cyrus Leung authored Jul 30, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  2ca5f82c
- [XPU] use `ZE_AFFINITY_MASK` for device select on xpu (#21815) · 05cbbe20
  Kunshang Ji authored Jul 30, 2025
```
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
```
  05cbbe20
- [Frontend] Add LLM.reward specific to reward models (#21720) · 65f311ce
  wang.yuqi authored Jul 30, 2025
```
Signed-off-by: wang.yuqi <noooop@126.com>
```
  65f311ce
- [Bugfix] Actually disable processing cache when API server is scaled out (#21839) · 44bc46da
  Cyrus Leung authored Jul 30, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  44bc46da
- [Bugfix] Fix comment typo of get_num_common_prefix_blocks() (#21827) · b7b23da4
  MingzhenHan authored Jul 30, 2025
```
Signed-off-by: MingzhenHan <hanmingzhen2002@outlook.com>
```
  b7b23da4
- [Bugfix] Fix shape mismatch assertion error when loading Gemma3n model with... · fdde1822
  Areeb Syed authored Jul 30, 2025
```
[Bugfix] Fix shape mismatch assertion error when loading Gemma3n model with BitsAndBytes quantization (#21808)
Signed-off-by: sydarb <areebsyed237@gmail.com>
```
  fdde1822
- Expose PyTorch profiler configuration to environment variables (#21803) · b917da44
  Csrayz authored Jul 30, 2025
```
Signed-off-by: Csrayz <33659823+Csrayz@users.noreply.github.com>
```
  b917da44
- [Docs] Switch to better markdown linting pre-commit hook (#21851) · ba5c5e54
  Harry Mellor authored Jul 30, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  ba5c5e54
- [v1][attention] Support Hybrid Allocator + FlashInfer (#21412) · 555e7225
  Chen Zhang authored Jul 29, 2025
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  555e7225
- [Bugfix] Correct max tokens for non-contiguous embeds (#21798) · 0e36abf9
  milesial authored Jul 29, 2025
```
Signed-off-by: Alexandre Milesi <30204471+milesial@users.noreply.github.com>
Co-authored-by: Alexandre Milesi <30204471+milesial@users.noreply.github.com>
```
  0e36abf9
29 Jul, 2025 13 commits
- [BugFix] Fix interleaved sliding window not set for Gemma3n (#21863) · 9266d980
  Yong Hoon Shin authored Jul 29, 2025
```
Signed-off-by: Yong Hoon Shin <yhshin@meta.com>
```
  9266d980
- docker: docker-aware precompiled wheel support (#21127) · a1873db2
  Doug Smith authored Jul 29, 2025
```
Signed-off-by: dougbtv <dosmith@redhat.com>
```
  a1873db2
- [Bugfix] Fix workspace buffer None issue for Flashinfer TRTLLM Backend (#21525) · 58b11b24
  elvischenv authored Jul 29, 2025
```
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
```
  58b11b24
- [Bugfix]fix mixed bits and visual language model quantization in AutoRound (#21802) · ad341c51
  Wenhua Cheng authored Jul 29, 2025
```
Signed-off-by: Wenhua Cheng <wenhua.cheng@intel.com>
```
  ad341c51
- [Bugfix] VLLM_V1 supports passing other compilation levels (#19340) · 04e38500
  Richard Zou authored Jul 29, 2025
```
Signed-off-by: Richard Zou <zou3519@gmail.com>
```
  04e38500
- [KVCache] Make KVCacheSpec hashable (#21791) · 755fa8b6
  Chen Zhang authored Jul 29, 2025
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  755fa8b6
- [Model] Refactor JambaForCausalLM (#21394) · 61a6905a
  Jee Jee Li authored Jul 29, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  61a6905a
- [V0 deprecation] Guided decoding (#21347) · 37efc63b
  Reza Barazesh authored Jul 29, 2025
```
Signed-off-by: Reza Barazesh <rezabarazesh@meta.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  37efc63b
- [Model]: Fused MoE for nomic-embed-text-v2-moe (#18321) · a4528f0c
  Isotr0py authored Jul 29, 2025
```
Signed-off-by: isotr0py <2037008807@qq.com>
Signed-off-by: Isotr0py <2037008807@qq.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  a4528f0c
- [Misc] Rework process titles (#21780) · 7234fe26
  Nick Hill authored Jul 29, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  7234fe26
- Migrate InternVLImageInputs and InternVLVideoInputs to TensorSchema (#21684) · f1e2c095
  Benji Beck authored Jul 28, 2025
```
Signed-off-by: Benji Beck <benjibeck@meta.com>
```
  f1e2c095
- skip fusedmoe layer for start_load_kv (#21378) · e18f0851
  Calvin Chen authored Jul 29, 2025
```
Signed-off-by: calvin chen <wen.chen@dynamia.ai>
```
  e18f0851
- [Refactor] Merge Compressed Tensor FP8 `CompressedTensorsW8A8Fp8MoEMethod` and... · 48b763d6
  Wentao Ye authored Jul 28, 2025
```
[Refactor] Merge Compressed Tensor FP8 `CompressedTensorsW8A8Fp8MoEMethod` and `CompressedTensorsW8A8Fp8MoECutlassMethod` (#21775)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  48b763d6
28 Jul, 2025 9 commits
- [Perf] Disable chunked local attention by default with llama4 (#21761) · 8aa1485f
  Lucas Wilkinson authored Jul 28, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  8aa1485f
- [Feat]: Add support for Dynamic Quant 4 bit CPU kleidiai kernels (#17112) · 89ac266b
  Nikhil Gupta authored Jul 28, 2025
```
Signed-off-by: Nikhil Gupta <nikhil.gupta2@arm.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
```
  89ac266b
- [Bugfix] DeepGEMM is not enabled on B200 due to `_lazy_init()` (#21472) · c6f36cfa
  Clayton Coleman authored Jul 28, 2025
```
Signed-off-by: Clayton Coleman <smarterclayton@gmail.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
```
  c6f36cfa
- Revert "[V1] Exception Handling when Loading KV Cache from Remote Store" (#21778) · b18b417f
  Kuntai Du authored Jul 28, 2025
```
Signed-off-by: KuntaiDu <kuntai@uchicago.edu>
```
  b18b417f
- [Bug] Enforce contiguous input for `dynamic_scaled_fp8_quant` and... · e0e58f97
  Wentao Ye authored Jul 28, 2025
```
[Bug] Enforce contiguous input for `dynamic_scaled_fp8_quant` and `static_scaled_fp8_quant` (#21773)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  e0e58f97
- [AMD][BugFix] Fix omission of wvSplitK kernel for small batch sizes (1-4) due... · b361f14e
  rasmith authored Jul 28, 2025
```
[AMD][BugFix] Fix omission  of wvSplitK kernel for small batch sizes (1-4) due to torch.compile (#21350)
Signed-off-by: Randall Smith <Randall.Smith@amd.com>
```
  b361f14e
- Use `metavar` to list the choices for a CLI arg when custom values are also accepted (#21760) · 94b71ae1
  Harry Mellor authored Jul 28, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  94b71ae1
- [P/D] Log warnings related to prefill KV expiry (#21753) · 7d44c691
  Nick Hill authored Jul 28, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  7d44c691
- [Bugfix] Fix granite speech shape validation (#21762) · e17a4d3b
  Cyrus Leung authored Jul 29, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  e17a4d3b