elvischenv f5228915a4

Merge branch 'main' into elvischenv/update-flashinfer

2025-12-23 09:33:11 +08:00

..

compressed_tensors

[Bugfix][ROCm] Fix typo: is_linear_fp8_enaled -> is_linear_fp8_enabled (#31109)

2025-12-21 21:14:58 -08:00

kernels

[Quantization] enable compressed-tensors marlin support for turing (#31000)

2025-12-18 20:29:48 -08:00

quark

[MoE Refactor][5/N] Isolate zero expert to LongCatFlash (#28891)

2025-12-20 18:22:04 +00:00

utils

[SM100] Enable fp8 compute for prefill MLA (#30746)

2025-12-22 19:15:57 +00:00

__init__.py

[CPU] Refactor CPU WNA16 (#28826)

2025-11-19 10:32:00 +08:00

auto_round.py

[Kernel][Quantization] add w4a8 support for marlin kernel (#24722)

2025-11-29 07:19:33 -08:00

awq_marlin.py

[MoE Refactor][5/N] Isolate zero expert to LongCatFlash (#28891)

2025-12-20 18:22:04 +00:00

awq_triton.py

…

awq.py

[Kernel][Quantization] add w4a8 support for marlin kernel (#24722)

2025-11-29 07:19:33 -08:00

base_config.py

…

bitblas.py

…

bitsandbytes.py

[MoE Refactor][5/N] Isolate zero expert to LongCatFlash (#28891)

2025-12-20 18:22:04 +00:00

cpu_wna16.py

[CPU] Refactor CPU WNA16 (#28826)

2025-11-19 10:32:00 +08:00

deepspeedfp.py

…

experts_int8.py

[MoE Refactor][5/N] Isolate zero expert to LongCatFlash (#28891)

2025-12-20 18:22:04 +00:00

fbgemm_fp8.py

…

fp8.py

[MoE Refactor][7/N] AITER MK (#31102)

2025-12-22 16:42:58 -07:00

fp_quant.py

…

gguf.py

[MoE Refactor][5/N] Isolate zero expert to LongCatFlash (#28891)

2025-12-20 18:22:04 +00:00

gptq_bitblas.py

…

gptq_marlin_24.py

…

gptq_marlin.py

[MoE Refactor][5/N] Isolate zero expert to LongCatFlash (#28891)

2025-12-20 18:22:04 +00:00

gptq.py

…

hqq_marlin.py

[Kernel][Quantization] add w4a8 support for marlin kernel (#24722)

2025-11-29 07:19:33 -08:00

inc.py

…

input_quant_fp8.py

[Bugfix][ROCm] Fix typo: is_linear_fp8_enaled -> is_linear_fp8_enabled (#31109)

2025-12-21 21:14:58 -08:00

ipex_quant.py

[XPU] enable fp8 online streaming quantization (#30944)

2025-12-20 13:45:27 +00:00

kv_cache.py

[Quantization] FP8 Weight Reloading for Quantized RL Rollout (#28480)

2025-12-09 13:54:32 -08:00

modelopt.py

[Feature]: Support NVIDIA ModelOpt HF FP8 variants FP8_PER_CHANNEL_PER_TOKEN and FP8_PB_WO in vLLM (#30957)

2025-12-21 22:34:49 -05:00

moe_wna16.py

[MoE Refactor][5/N] Isolate zero expert to LongCatFlash (#28891)

2025-12-20 18:22:04 +00:00

mxfp4.py

Merge branch 'main' into elvischenv/update-flashinfer

2025-12-22 10:55:04 -08:00

petit.py

[Attention] Update attention imports (#29540)

2025-11-27 11:19:09 -05:00

ptpc_fp8.py

[Attention] Update attention imports (#29540)

2025-11-27 11:19:09 -05:00

qutlass_utils.py

…

rtn.py

[MoE Refactor][5/N] Isolate zero expert to LongCatFlash (#28891)

2025-12-20 18:22:04 +00:00

schema.py

…

torchao.py

Support using Int4PreshuffledTensor after loading (#26066)

2025-11-04 06:00:57 -05:00

tpu_int8.py

…