Jiangyun Zhu 285bab4752 [Kernel] use flashinfer for gdn prefill (#32846) ...

Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>

2026-02-09 12:17:25 -05:00

..

assets

…

benchmarks

[ASR] Fix audio benchmark and add RTFx metric (#32300)

2026-02-09 10:02:37 +00:00

compilation

[Kernel] FlashInfer: switch allreduce fusion to unified API (#33985)

2026-02-09 15:43:24 +00:00

config

[UX] Add --language-model-only for hybrid models (#34120)

2026-02-09 14:57:33 +00:00

device_allocator

…

distributed

Add NUMA Core binding in nixl_connector for CPU xPyD (#32365)

2026-02-09 15:39:12 +00:00

engine

[UX] Add --language-model-only for hybrid models (#34120)

2026-02-09 14:57:33 +00:00

entrypoints

[Frontend][last/5] Make pooling entrypoints request schema consensus. (#31127)

2026-02-09 06:42:38 +00:00

grpc

…

inputs

…

kernels

…

logging_utils

…

lora

[Tiny] Rename encoder budget file to more specific name (#34103)

2026-02-09 03:48:19 +00:00

model_executor

[Kernel] use flashinfer for gdn prefill (#32846)

2026-02-09 12:17:25 -05:00

multimodal

[Tiny] Rename encoder budget file to more specific name (#34103)

2026-02-09 03:48:19 +00:00

parser

…

platforms

Add NUMA Core binding in nixl_connector for CPU xPyD (#32365)

2026-02-09 15:39:12 +00:00

plugins

…

profiler

…

ray

…

reasoning

…

renderers

…

third_party

…

tokenizers

…

tool_parsers

…

tracing

…

transformers_utils

[MODEL] Adding Support for Qwen3.5 Models (#34110)

2026-02-09 21:12:58 +08:00

triton_utils

…

usage

…

utils

[Frontend][last/5] Make pooling entrypoints request schema consensus. (#31127)

2026-02-09 06:42:38 +00:00

v1

Add NUMA Core binding in nixl_connector for CPU xPyD (#32365)

2026-02-09 15:39:12 +00:00

vllm_flash_attn

…

__init__.py

…

_aiter_ops.py

…

_bc_linter.py

…

_custom_ops.py

…

_ipex_ops.py

…

beam_search.py

…

collect_env.py

…

connections.py

…

env_override.py

…

envs.py

[Release 2.10] Update to Torch 2.10 - final release (#30525)

2026-02-08 13:51:09 -08:00

exceptions.py

…

forward_context.py

[torch.compile] Add an option to force-enable the MOE cold start optimization (#33735)

2026-02-08 18:42:56 +00:00

logger.py

…

logits_process.py

…

logprobs.py

…

model_inspection.py

…

outputs.py

…

pooling_params.py

…

py.typed

…

sampling_params.py

…

scalar_type.py

…

scripts.py

…

sequence.py

…

tasks.py

…

version.py

…