Yuan Luoandluoyuan.luo b9af8d2eb9 [VLM] Support apply qk norm in multi cuda streams (#15720) ...

Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>

2025-12-25 14:35:07 +08:00

..

__init__.py

…

aio_rwlock.py

…

bench_utils.py

…

common.py

[NPU] Bug fix in device detect (#14137)

2025-12-25 00:18:39 +08:00

cuda_ipc_transport_utils.py

unified management of environment variables for vlm cuda ipc transport (#14501)

2025-12-18 12:28:06 +08:00

device_timer.py

Add metrics for having prefill and decode in different ranks (#15752)

2025-12-24 21:35:35 +08:00

hf_transformers_utils.py

Monkey patch deepseek-ocr's v_head_dim (#15384)

2025-12-18 16:28:07 +08:00

host_shared_memory.py

…

mistral_utils.py

Mistral Large 3 NVFP4 TRTLLM MoE support (#15049)

2025-12-18 11:11:42 +08:00

multi_stream_utils.py

[VLM] Support apply qk norm in multi cuda streams (#15720)

2025-12-25 14:35:07 +08:00

numa_utils.py

…

nvtx_pytorch_hooks.py

…

offloader.py

…

patch_torch.py

[NPU]Fix for ipc handle with npu (#14138)

2025-12-19 22:39:04 +08:00

poll_based_barrier.py

…

profile_merger.py

…

profile_utils.py

…

request_logger.py

Support JSON format request logging for easier parsing (#15743)

2025-12-24 19:52:43 +08:00

rpd_utils.py

…

slow_rank_detector.py

…

torch_memory_saver_adapter.py

…

watchdog.py

Tiny add stuck simulation (#15613)

2025-12-22 17:00:18 +08:00

weight_checker.py

…