intel/auto-round

Pull Requests Commits

support vllm based model quantization

changwangss committed 7 hours ago

6939e027

[ARK] Add varlen support for sdpa and update tests (#1957)

luoyu-intel committed 1 day ago

Verified 948c7783

Adapt the latest sglang integration test Engine kwargs. (#1964)

changwangss committed 2 days ago

Verified 49244549

fix auto_scheme gguf format wrong bits if use tiny model (#1953)

n1ck-guo committed 2 days ago

Verified b1a6baf9

speed up awq ci (#1954)

WeiweiZhang1 committed 5 days ago

Verified 9468e52c

add other non-integer scheme support for awq (#1892)

WeiweiZhang1 committed 5 days ago

Verified 5a3ade8c

Remove kernels version limit in test (#1945)

chensuyue committed 6 days ago

Verified 142e91b4

Use shared tokenizer save helper in exporters (#1951)

changwangss committed 6 days ago

Verified 436b9db6

update ARK README (#1922)

luoyu-intel committed 6 days ago

Verified f42a0421

Implement shard parallelism configuration and update documentation (#1942)

xin3he committed 7 days ago

Verified 7e2b69d4

fix saving hadamard rotation model without rotation config. (#1933)

lkk12014402 committed 7 days ago

Verified 6a2e5ad3

fix few gguf bugs of hf-name and checkpoint name (#1939)

n1ck-guo committed 7 days ago

Verified 126793d5

Add flashinfer-jit-cache installation to cuda unit test setup (#1947)

XuehaoSun committed 8 days ago

Verified 349b0418

fix issue of step3p7 and CI failure (#1929)

xin3he committed 8 days ago

Verified 1c4baa03

Update PyTorch and torchvision versions to 2.12.1 and 0.27.1 in CI (#1943)

XuehaoSun committed 8 days ago

Verified 354c6a3b

Update CPU agent pool (#1936)

chensuyue committed 8 days ago

Verified 35650a09

Fix SIGSEGV in pack-weight unpack/GEMM on malformed blob (#1928)

luoyu-intel committed 9 days ago

Verified 495e2539

Refine AutoScheme logic for gguf, reduce memory consumption and improve speed (#1916)

wenhuach21 committed 9 days ago

Verified 0c9c5b1d

DDP for multiple cards (#1897)

yiliu30 committed 11 days ago

Verified 33b7df0e

reorder UT for CUDA CI memory leak (#1938)

xin3he committed 13 days ago

Verified 0cc47c69

update what's new: vllm-omni integrates AutoROund (#1931)

xin3he committed 13 days ago

Verified 759a9ad7

Enhance b70 check (#1937)

yiliu30 committed 13 days ago

Verified e8709167

Support Wan2.2 S2V Quantization (#1924)

changwangss committed 13 days ago

Verified ca378799

Fix UT coverage (#1927)

XuehaoSun committed 13 days ago

Verified 6b311875

[Model Free] Add MXFP mix-precision quantization and support RoutedExperts (#1921)

xin3he committed 14 days ago

Verified be043f32

refactor and support for multi algs fusion (#1920)

n1ck-guo committed 14 days ago

Verified df9299da

enable MXFP clamp for model free (#1914)

xin3he committed 19 days ago

Verified 50ee58c9

fix awq cuda CI (#1912)

WeiweiZhang1 committed 20 days ago

Verified 34790de7

fix inplace rotation issue (#1903)

wenhuach21 committed 20 days ago

Verified 9d5edd56

[ARK] update README (#1906)

luoyu-intel committed 20 days ago

Verified c014b0eb

Older