Vadim Gimpelson 4ff8c3c8f9 [BUGFIX][Mamba][Qwen3.5] Zero freed SSM cache blocks on GPU (#35219) ...

Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>

2026-03-10 03:32:20 -07:00

..

__init__.py

…

argparse_utils.py

Enforce that model is the first positional arg when --served-model-name is used (#34973)

2026-02-23 08:38:05 -08:00

async_utils.py

…

cache.py

…

collection_utils.py

[Plugin] Simplify IO Processor Plugin interface (#34236)

2026-02-10 19:47:39 -08:00

counter.py

…

deep_gemm.py

[Feat] Add CUDA torch fallbacks for fp8_mqa_logits/fp8_paged_mqa_logits_torch function (#35271)

2026-02-28 10:12:00 +00:00

flashinfer.py

[MoE Refactor] Create MK for TRTLLM Kernels (#32564)

2026-03-03 10:39:50 -08:00

func_utils.py

…

gc_utils.py

…

hashing.py

…

import_utils.py

[WideEP] Remove pplx all2all backend (#33724)

2026-02-26 14:30:10 -08:00

jsontree.py

…

math_utils.py

[BUGFIX][Mamba][Qwen3.5] Zero freed SSM cache blocks on GPU (#35219)

2026-03-10 03:32:20 -07:00

mem_constants.py

…

mem_utils.py

[Hardware] Replace torch.cuda.empty_cache with torch.accelerator.empty_cache (#30681)

2026-03-04 09:49:47 +00:00

mistral.py

[Feature] Lazy import for the "mistral" tokenizer module. (#34651)

2026-02-23 00:43:01 -08:00

nccl.py

…

network_utils.py

[BugFix] avoid infinite loop with VLLM_PORT and get_open_ports_list (#36191)

2026-03-05 22:15:29 -08:00

nvtx_pytorch_hooks.py

…

platform_utils.py

[Platform] Add current_platform.num_compute_units interface (#35042)

2026-02-24 22:22:49 -08:00

print_utils.py

[Examples][1/n] Resettle basic examples. (#35579)

2026-03-08 20:22:53 -07:00

profiling.py

…

registry.py

…

serial_utils.py

…

system_utils.py

[ROCm] Derive device capability from GCN arch string without CUDA init (#35069)

2026-02-28 13:55:28 +08:00

tensor_schema.py

…

torch_utils.py

[torch.compile] Undo the fast_moe_cold_start hack in torch>=2.11 (#35475)

2026-03-01 21:44:22 +00:00

tqdm_utils.py

[Renderer] Move InputPreprocessor into Renderer (2/2) (#34560)

2026-02-17 05:29:01 -08:00