Commits · 47512b3200156cc0db4a62791d2f7fefcaecb6ad · OpenDAS / vllm_cscc

"tests/compile/__init__.py" did not exist on "376725ce74d2d75490eed1840b41de00c0e4acd6"

08 Mar, 2025 7 commits
- Default to `generation_config` from model (#12622) · 47512b32
  Harry Mellor authored Mar 08, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  47512b32
- [core] add `extra_args` to `SamplingParams` (#13300) · 4aae6676
  Aviv Keshet authored Mar 07, 2025
```
Signed-off-by: Aviv Keshet <akeshet@scaledcognition.com>
```
  4aae6676
- [MISC][V1] Register process killing handler only in the main thread (#14380) · 9f3bc0f5
  Cody Yu authored Mar 07, 2025
```
Signed-off-by: Cody Yu <hao.yu.cody@gmail.com>
```
  9f3bc0f5
- [Bugfix][Disaggregated] Add a check in send_kv_caches_and_hidden_states and... · 980385f8
  Mathis Felardos authored Mar 08, 2025
```
[Bugfix][Disaggregated] Add a check in send_kv_caches_and_hidden_states and fix the reshape of the KVCache (#14369)
Signed-off-by: Mathis Felardos <mathis@mistral.ai>
```
  980385f8
- Revert "[Perf] Reduce MLA CPU overheads in V1 (#14384)" (#14471) · ca7a2d5f
  Tyler Michael Smith authored Mar 08, 2025
  
  ca7a2d5f
- [Bugfix][V1] Handle MLA in kv_cache_interface (#14462) · 33368140
  Tyler Michael Smith authored Mar 08, 2025
```
Signed-off-by: Tyler Michael Smith <tyler@neuralmagic.com>
```
  33368140
- [V1] Prompt logprobs + APC compatibility; prompt logprobs reqs cannot fill APC (#13949) · ef640440
  afeldman-nm authored Mar 07, 2025
  
  ef640440
07 Mar, 2025 15 commits
- [Bugfix] Fix torch_xla which can't handle None seed introduced in #14274 (#14459) · 66e16a03
  yarongmu-google authored Mar 07, 2025
```
Signed-off-by: Yarong Mu <ymu@google.com>
```
  66e16a03
- [V1][Metrics] Fix traceback with preemptions+LoRA (#14220) · e1f0835a
  Mark McLoughlin authored Mar 07, 2025
```
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
```
  e1f0835a
- [V1] Eagerly remove finished requests from the batch (#14388) · 8ed5421a
  Nick Hill authored Mar 07, 2025
```
Signed-off-by: Nick Hill <nhill@redhat.com>
```
  8ed5421a
- [v1] torch.compile integration explanation (#14437) · c6359e8c
  youkaichao authored Mar 08, 2025
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
```
  c6359e8c
- [Misc] Add Phi4-MM example (#14343) · 952a0749
  Jee Jee Li authored Mar 08, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  952a0749
- [Kernel] optimize performance of gptq marlin kernel when n is small (#14138) · d0feea31
  Jinzhen Lin authored Mar 08, 2025
```
Signed-off-by: Jinzhen Lin <linjinzhen@hotmail.com>
```
  d0feea31
- [V1][Core] Support for Structured Outputs (#12388) · 80e9afb5
  Aaron Pham authored Mar 07, 2025
```
Signed-off-by: Aaron Pham <contact@aarnphm.xyz>
Signed-off-by: Russell Bryant <rbryant@redhat.com>
Co-authored-by: Russell Bryant <rbryant@redhat.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
```
  80e9afb5
- Use the optimized block sizes after tuning the kernel. (#14329) · 1e3598ed
  iefgnoix authored Mar 07, 2025
  
  1e3598ed
- Fix missing `kv_caches` and `attn_metadata` in `OpenVINOCausalLM` (#14271) · f7a6bd0f
  Harry Mellor authored Mar 07, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  f7a6bd0f
- [BUGFIX] Skip tokenization support for throughput benchmark (#12712) · 0ca3b8e0
  Aleksandr Malyshev authored Mar 07, 2025
```
Signed-off-by: root <root@banff-cyxtera-s73-5.ctr.dcgpu>
Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: root <root@banff-cyxtera-s73-5.ctr.dcgpu>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
```
  0ca3b8e0
- [Misc] Set default value of seed to None (#14274) · cc102814
  மனோஜ்குமார் பழனிச்சாமி authored Mar 07, 2025
```
Signed-off-by: மனோஜ்குமார் பழனிச்சாமி <smartmanoj42857@gmail.com>
```
  cc102814
- [Bugfix] Clean up multi-modal processors (#14417) · 05fb6718
  Cyrus Leung authored Mar 07, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  05fb6718
- [Bugfix] Fix JambaForCausalLM LoRA (#14370) · ddd1ef66
  Jee Jee Li authored Mar 07, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  ddd1ef66
- [FP8] Refactor apply_fp8_linear and apply_fp8_linear_generic into an object (#14390) · e1744502
  Luka Govedič authored Mar 07, 2025
```
Signed-off-by: luka <luka@neuralmagic.com>
```
  e1744502
- [Perf] Reduce MLA CPU overheads in V1 (#14384) · dae68969
  Lucas Wilkinson authored Mar 06, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  dae68969
06 Mar, 2025 18 commits
- [Hardware][TPU]Enable ragged paged attention kernel and resolve recompilation issue (#14310) · 0578e5a4
  Chengji Yao authored Mar 06, 2025
```
Signed-off-by: Chengji Yao <chengjiyao@google.com>
```
  0578e5a4
- [V1][Bugfix] Standardize quantized kv cache rejection for attention backends (#14221) · 6832707e
  Michael Goin authored Mar 06, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  6832707e
- [Bug] Fix Attention when ignored in by quant_method (#14313) · 6b2ef5cd
  Michael Goin authored Mar 06, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  6b2ef5cd
- [Bugfix] Fix use_direct_call condition in FusedMoE layer for (#14382) · 958adce4
  Tyler Michael Smith authored Mar 06, 2025
```
Signed-off-by: Tyler Michael Smith <tyler@neuralmagic.com>
```
  958adce4
- [Distributed] Add enable_expert_parallel arg (#14305) · cc2f9b32
  Tyler Michael Smith authored Mar 06, 2025
```
Signed-off-by: Tyler Michael Smith <tyler@neuralmagic.com>
```
  cc2f9b32
- [V1] Do not detokenize if sampling param detokenize is False (#14224) · cd579352
  Himanshu Jaju authored Mar 06, 2025
```
Signed-off-by: Himanshu Jaju <hj@mistral.ai>
Signed-off-by: Nick Hill <nhill@redhat.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
```
  cd579352
- Fix mla prefill context performance (#13897) · 9f1710f1
  Ying Zhong authored Mar 07, 2025
```
Signed-off-by: ZhongYingMatrix <zhongyingmatrix@gmail.com>
```
  9f1710f1
- Add authors to license header. (#14371) · e642ec96
  Thomas Parnell authored Mar 06, 2025
```
Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>
Co-authored-by: Burkhard Ringlein <ngl@zurich.ibm.com>
Co-authored-by: Jan van Lunteren <jvl@zurich.ibm.com>
```
  e642ec96
- Reinstate `best_of` for V0 (#14356) · bf0560bd
  Harry Mellor authored Mar 06, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  bf0560bd
- [RLHF] use worker_extension_cls for compatibility with V0 and V1 (#14185) · 151b08e0
  youkaichao authored Mar 07, 2025
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
```
  151b08e0
- [Core] Don't use cache during multi-modal profiling (#14336) · 82551ad6
  Cyrus Leung authored Mar 07, 2025
  
  82551ad6
- [Bugfix][Core] fix abort_seq_group and memory leak when n>1 (#14326) · caac5c2e
  courage17340 authored Mar 06, 2025
```
Signed-off-by: courage17340 <courage17340@163.com>
```
  caac5c2e
- [Kernel] [V1] Improved performance for V1 Triton (ROCm) backend (#14152) · 6bd1dd9d
  Thomas Parnell authored Mar 06, 2025
  
  6bd1dd9d
- [Frontend][Docs] Transcription API streaming (#13301) · fa82b938
  Nicolò Lucchesi authored Mar 06, 2025
```
Signed-off-by: NickLucche <nlucches@redhat.com>
```
  fa82b938
- [Core] Optimizing cross-attention `QKVParallelLinear` computation (#12325) · 69ff99fd
  Nicolò Lucchesi authored Mar 06, 2025
```
Signed-off-by: NickLucche <nlucches@redhat.com>
Signed-off-by: NickLucche <nick@nlucches-4xa100.c.openshift-330514.internal>
Co-authored-by: NickLucche <nick@nlucches-4xa100.c.openshift-330514.internal>
```
  69ff99fd
- [V1][VLM][Pixtral-HF] Support Pixtral-HF on V1 (#14275) · 5d802522
  lkchen authored Mar 06, 2025
```
Signed-off-by: Linkun Chen <github@lkchen.net>
```
  5d802522
- [Model] Update Paligemma multimodal processing with PromptUpdate (#14015) · 17699280
  kYLe authored Mar 06, 2025
```
Signed-off-by: Kyle Huang <kylhuang@nvidia.com>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
```
  17699280
- [Hardware] Update the flash attn tag to support Blackwell (#14244) · ed6ea065
  Pavani Majety authored Mar 05, 2025
  
  ed6ea065