intel/auto-round

Pull Requests Commits

add round_whisper method and accuracy script

sys-lpot-val committed 196 days ago

2a6e3acb

xin3he committed 199 days ago

c124ae7f

support accuracy verification

xin3he committed 199 days ago

aade3185

support autoround calibration

xin3he committed 206 days ago

7c1bc200

Adding a workspace for reproducing

xin3he committed 206 days ago

2cbb12fd

add llmcompressor format

xin3he committed 208 days ago

c2118215

add torch backend (#555)

WeiweiZhang1 committed 210 days ago

Verified ce8eedd7

switch to transformers API in cpu ut (#580)

wenhuach21 committed 210 days ago

Verified 6ceaa8af

improve the robustness of parsing vlm config (#577)

wenhuach21 committed 210 days ago

Verified 69749b91

modify default asym 4bits auto-round format to awq, fix save folder typo for mllm (#575)

WeiweiZhang1 committed 211 days ago

Verified c1c8148e

add sym & asym gguf quant for gguf baseline (iter==0) (#573)

n1ck-guo committed 211 days ago

Verified fde99cc8

add gguf accuracy data (#574)

wenhuach21 committed 212 days ago

Verified 3e79ebcf

support for llava-based hf model (#568)

n1ck-guo committed 215 days ago

Verified 91487375

update readme (#571)

wenhuach21 committed 216 days ago

Verified 1bad5ee2

update readme (#569)

wenhuach21 committed 216 days ago

Verified 90fef628

Support for more gguf format and float zp for Q*_1 (#560)

n1ck-guo committed 216 days ago

Verified b38d19dc

update habana docker to 1.21.0 (#566)

XuehaoSun committed 217 days ago

Verified 9a23061b

fix typo in backend (#564)

wenhuach21 committed 222 days ago

Verified eee6db4b

support xpu in triton backend (#563)

wenhuach21 committed 222 days ago

Verified f683f99a

add UE5M3 simulation (#562)

wenhuach21 committed 223 days ago

Verified fa41c3d1

refine xpu requirement/config json and fix several issues (#558)

wenhuach21 committed 226 days ago

Verified a45eadb3

[GGUF STEP4] fix search bug and improve packing & eval speed (#545)

n1ck-guo committed 230 days ago

Verified 4b85b512

itrex requires torch<2.7 (#548)

XuehaoSun committed 238 days ago

Verified 2e842a2f

add recipes for qwen3 8b and 14b (#552)

n1ck-guo committed 239 days ago

Verified 6a6b5901

update readme (#550)

wenhuach21 committed 240 days ago

Verified 59cf22ed

[critic bug]remove redundant round in dq simulation (#543)

wenhuach21 committed 243 days ago

Verified c6155fdb

fix_quantization_config_for_inference (#542)

WeiweiZhang1 committed 244 days ago

Verified 76576406

freeze pytorch & ipex version in CI (#541)

XuehaoSun committed 244 days ago

Verified 784273ed

bump version into v0.5.1 (#540)

XuehaoSun committed 245 days ago

Verified 336ef66a

fix triton multiple gpus and some other issues (#539)

wenhuach21 committed 246 days ago

Verified d6d6dad1

Older