Stefano CastagnettaandKartikey Vyas 4e0681ed56 [Quantization] Support packed NVFP4 Qwen4Exp PLE embeddings (#56273) ...

Signed-off-by: Stefano Castagnetta <scastagnetta@nvidia.com>
Co-authored-by: Kartikey Vyas <kartikey@artificialanalysis.ai>

2026-10-14 08:24:55 +08:00

..

configs

[XPU][Kernel] Tune Triton W8A8 block-FP8 GEMM for Intel B70 (#56063)

2026-10-02 12:56:11 +01:00

humming

[Bugfix][Quantization] Accept folded NVFP4 scales for Humming A16 (#60337)

2026-10-07 11:04:43 -07:00

__init__.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

b12x_moe.py

[Kernel] Add b12x FP4 MoE backend (#52018)

2026-08-21 08:04:36 -07:00

config_utils.py

[Quantization] Enable shared expert fusion compatibility with online shared_expert quantization (showcase: along Quark MXFP4 routed experts) (#55686)

2026-10-08 12:04:46 -04:00

flashinfer_fp4_moe.py

[quantization] Add per-token NVFP4 quantization MoE support for ReLU2 (#56740)

2026-10-07 22:01:09 +00:00

flashinfer_mxint4_moe.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

flashinfer_utils.py

[quantization] Add per-token NVFP4 quantization MoE support for ReLU2 (#56740)

2026-10-07 22:01:09 +00:00

fp8_utils.py

[Feature] Support batch invariance on ROCm (#52231)

2026-10-12 10:53:26 -05:00

gptq_utils.py

[Bugfix][Quant] Read safetensors headers of checkpoints with other file names (#60411)

2026-10-08 01:38:41 -05:00

int8_utils.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

layer_utils.py

Update Optional[x] -> x | None and Union[x, y] to x | y (#26633)

2025-10-12 09:51:31 -07:00

machete_utils.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

marlin_utils_fp4.py

[Core][Kernel] Share persistent workspaces for Marlin and Humming (#57421)

2026-09-20 17:28:33 +08:00

marlin_utils_fp8.py

[Core][Kernel] Share persistent workspaces for Marlin and Humming (#57421)

2026-09-20 17:28:33 +08:00

marlin_utils_test.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

marlin_utils.py

[Core][Kernel] Share persistent workspaces for Marlin and Humming (#57421)

2026-09-20 17:28:33 +08:00

mxfp4_utils.py

[Kernel] Enable OAI Triton MXFP4 MoE on SM12x (#58877)

2026-10-06 01:49:13 -04:00

mxfp6_utils.py

[Chore] Clean up pytorch helper functions in vllm.utils (#26908)

2025-10-18 09:48:22 -07:00

mxfp8_utils.py

[Model] Support DeepSeek-V4.1-Flash (#56214)

2026-09-11 02:11:19 -07:00

nvfp4_emulation_utils.py

[Quantization] Support packed NVFP4 Qwen4Exp PLE embeddings (#56273)

2026-10-14 08:24:55 +08:00

nvfp4_utils.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

ocp_mx_utils.py

[Bugfix][ROCm] AMD-Quark mixed-precision DeepSeek-V4.1 support (#57071)

2026-09-25 20:51:18 -05:00

quant_utils.py

[feat] FlashInfer CuteDSL MegaMoE integration (#54049)

2026-10-02 15:43:33 -04:00

w8a8_utils.py

[Refactor] Remove dead code quantization 2 (#52821)

2026-08-25 20:28:44 -07:00