[Bugfix] Fix FP16 overflow for DeepSeek V2 (#13232)

Concurrensee committed over 1 year ago

[V1][PP] Do not block engine core when no requests to schedule (#14585)

Cody Yu committed over 1 year ago

[neuron] add reshape_and_cache (#14391)

Liangfu Chen committed over 1 year ago

[Neuron] Add Neuron device communicator for vLLM v1 (#14085)

gnovack committed over 1 year ago

[MISC][V1] Handle exception of current_platform.get_device_name() in arg_utils (#14379)

Cody Yu committed over 1 year ago

[Kernel] moe wna16 cuda kernel (#13321)

Jinzhen Lin committed over 1 year ago

[V1] Prevent xgrammar from breaking TPU support (#14575)

Russell Bryant committed over 1 year ago

Fix typo in benchmark_serving_structured_output.py (#14566)

Russell Bryant committed over 1 year ago

[V1] LoRA - Add triton kernels for V1 (#13096)

Varun Sundar Rabindranath committed over 1 year ago

[Minor] Update the tqdm bar for parallel sampling (#14571)

Woosuk Kwon committed over 1 year ago

[Perf] Improve MLA on V1 (#14540)

Simon Mo committed over 1 year ago

[Bugfix][v1] fixed llava-hf/llava-1.5-7b-hf is broken on V1 (#14554)

Chauncey committed over 1 year ago

[Docs] Make installation URLs nicer (#14556)

Harry Mellor committed over 1 year ago

Correct capitalisation: `VLLM` -> `vLLM` (#14562)

Harry Mellor committed over 1 year ago

[V1][Bugfix] Fix handing of `second_per_grid_ts` for Qwen2-VL & Qwen2.5-VL (#14548)

Roger Wang committed over 1 year ago

Correct capitalisation: `Github` -> `GitHub` (#14561)

Harry Mellor committed over 1 year ago

[Doc] Update PaliGemma note to a warning (#14565)

Cyrus Leung committed over 1 year ago

[Kernel] Add more dtype support for GGUF kernels (#14043)

Szymon Ożóg committed over 1 year ago

[Frontend] support image embeds (#13955)

Chauncey committed over 1 year ago

[Docs] Mention `model_impl` arg when explaining Transformers fallback (#14552)

Harry Mellor committed over 1 year ago