intel/auto-round

Pull Requests Commits

n1ck-guo committed 165 days ago

4e9b5aeb

[STEP 2] split compressor into few quantizers

n1ck-guo committed 165 days ago

bfd6882d

add support for scheme FP8_STATIC to export llm_compressor format (#816)

n1ck-guo committed 168 days ago

Verified 1089004a

[STEP 1] merge all api(MLLM, Adam) into AutoRound (#791)

n1ck-guo committed 168 days ago

Verified bd4bc2c9

bump version to 0.8.0.dev (#830)

XuehaoSun committed 169 days ago

Verified 95973a4f

move environment variable setting into eval function (#829)

xin3he committed 169 days ago

Verified 2a1409db

rm triton from requirements.txt (#824)

wenhuach21 committed 171 days ago

Verified c898d132

enhance auto-round eval with vllm backend (#815)

xin3he committed 172 days ago

Verified d74f4bcf

Fix llm-compressor export (#820)

Kaihui-intel committed 175 days ago

Verified 40aed064

Refine logger and add envs (#817)

yiliu30 committed 176 days ago

Verified c5302e16

fix nsamples in get_dataloader (#804)

wenhuach21 committed 176 days ago

Verified 00fff814

Bump release version to v0.7.0 (#814)

XuehaoSun committed 177 days ago

Verified 30d1061f

fix gptqmodel inference issue (#813)

wenhuach21 committed 177 days ago

Verified c4e7fbe5

fix bug of q_layer_inputs (#811)

n1ck-guo committed 177 days ago

Verified e1ae6e93

Support loading for static quant weight fp8 act fp8 (#730)

yiliu30 committed 178 days ago

Verified 09e4d312

remove torch compile (#807)

wenhuach21 committed 178 days ago

Verified 1452d95f

fix cuda ut bug of use_deterministic_algorithms (#805)

n1ck-guo committed 178 days ago

Verified 886d6571

fix mxfp exporting (#806)

WeiweiZhang1 committed 178 days ago

Verified 520c78fe

improve the robustness of scheme (#803)

wenhuach21 committed 178 days ago

Verified 9534d2ac

AutoRoundMLLM supports scheme and fix device_map=dict regression (#801)

n1ck-guo committed 179 days ago

Verified 78fce450

enable model python files saving (#802)

WeiweiZhang1 committed 179 days ago

Verified da089e9f

fix several cuda ut bug (#797)

n1ck-guo committed 179 days ago

Verified ab55a973

fix typo in readme (#799)

wenhuach21 committed 179 days ago

Verified e39f25e1

refine readme (#798)

wenhuach21 committed 179 days ago

Verified bc045f93

fix several regressions including lm-head quantization, 3bit asym torch backend,etc (#796)

wenhuach21 committed 179 days ago

Verified 08d6b9e7

add fp8 export format check (#779)

n1ck-guo committed 179 days ago

Verified 5ead4c6d

align formats packing device to API (#795)

WeiweiZhang1 committed 182 days ago

Verified dc81a878

rename llmcompressor to llm_compressor for align with other formats (#780)

WeiweiZhang1 committed 182 days ago

Verified e66652a4

fix bug and add nvfp in alg-ext with slight improvement (#794)

wenhuach21 committed 182 days ago

Verified 7e014ca3

Support auto device mapping (#781)

Kaihui-intel committed 182 days ago

Verified d7d2efad

Older