[Core][DSV4] Skip caching SWA blocks that can never serve a prefix-cache hit (#42258)

Yifan Qiao committed 6 months ago

[Misc] Make it simpler to replace out-of-tree layer classes with related LoRA layers. (#42306)

yupeng committed 6 months ago

[Deprecation] Remove old locations of `get_tokenizer` and `resolve_hf_chat_template` (#35024)

Cyrus Leung committed 6 months ago

[Bugfix] Clarify CPU backend memory error messages reference shared flag (#42479)

Daniel W committed 6 months ago

Update Intel Xeon model list and vLLM Benchmark Suite BKMs (#42607)

Louie Tsai committed 6 months ago

[Bugfix] Fix inverted condition causing thinking_token_budget to be silently ignored (#41674)

Keyi Li committed 6 months ago

[Model] Support InternS2 Preview (#42705)

Isotr0py committed 6 months ago

[CPU][RISC-V] Add RVV-optimized attention kernels for RISC-V Vector Extension (#40119)

lyd1992 committed 6 months ago

[Bugfix] Fix incorrect chat template format for Qwen3.5 (#42660)

Cyrus Leung committed 6 months ago

Bump llguidance to 1.7 (#42150)

Chao-Ju Chen committed 6 months ago

[Quant] Consolidate GPTQ: rename gptq_marlin.py to auto_gptq.py (#38288)

Chengyi Nie committed 6 months ago

[Model Runner V2][Bug Fix][DSV4] Ensure lazy attention state initializations happen during cudagraph capture (#42444)

Giancarlo Delfin committed 6 months ago

[Bugfix] Add swiglu limits to deepgemm fp8 methods (#41986)

Yongye Zhu committed 6 months ago

[CI][ROCm] Remove unsupported cases in test_fusion.py (#38680)

Charlie Fu committed 6 months ago

[Aiter][ROCm] RMSNormGated+GroupedQuantFP8 fusion (#40710)

Tres committed 6 months ago

[Attention] Remove deprecated MLA prefill arguments (#42555)

Matthew Bonanni committed 6 months ago

[Model Runner v2] Oracle for model runner v2 - qwen3 dense model by default [1/N] (#39337)

Wentao Ye committed 6 months ago

[Bugfix][Multimodal] PyAV video backend returns keyframes labeled as targets (#42586)

Ranran committed 6 months ago

[Fix] Misc Fixes in ViT CUDA Graph (#38040)

Baorun (Lauren) Mu committed 6 months ago

[Bugfix] Fix LM detection for Nemotron Parse (#42641)

Cyrus Leung committed 6 months ago