intel/auto-round

Pull Requests Commits

xinhe3 committed 164 days ago

f91f480d

xinhe3 committed 169 days ago

fa358e22

xinhe3 committed 175 days ago

b66f785c

[pre-commit.ci] auto fixes from pre-commit.com hooks

pre-commit-ci[bot] committed 176 days ago

5703ff37

Merge branch 'main' into xinhe/deepspeed

xin3he committed 176 days ago

Verified 23bd8a9a

Update auto_round/wrapper.py

xin3he committed 176 days ago

Verified 7fea2d98

Update auto_round/utils.py

xin3he committed 176 days ago

Verified 5edf3d69

support deepspeed LinearLayer and LinearAllreduce

xinhe3 committed 176 days ago

d2a1fb40

support for static activation quantization calibration with group_size (#693)

n1ck-guo committed 178 days ago

Verified 6afe07cb

support to save image_processor (#694)

n1ck-guo committed 178 days ago

Verified 4079a810

Add script for cuda unit test (#567)

XuehaoSun committed 178 days ago

Verified 4e617551

Update readme (#695)

wenhuach21 committed 178 days ago

Verified a598f17a

support static global scale for nvfp4 and update readme (#691)

wenhuach21 committed 178 days ago

Verified 4a3530fe

Add black and blacken-docs in pre-commit (#692)

XuehaoSun committed 181 days ago

Verified d424d3b4

support rceil for mxfp (#660)

wenhuach21 committed 182 days ago

Verified bec58a23

better log (#688)

wenhuach21 committed 182 days ago

Verified c6bde4b4

Add ruff and isort (#578)

XuehaoSun committed 182 days ago

Verified 774a5ae0

support to export static afp8 model (#662)

n1ck-guo committed 182 days ago

Verified 92bbaad3

fix ut import issue (#686)

WeiweiZhang1 committed 183 days ago

Verified 9e475278

bump version into v0.6.0 (#687)

XuehaoSun committed 183 days ago

Verified 5a07e4a7

fix gemma vlm gguf regression (#685)

wenhuach21 committed 183 days ago

Verified 3f36254c

fix gguf regression for large models (#680)

wenhuach21 committed 184 days ago

Verified 21bb06b6

update gguf convert file and fix bug of permute bug (#679)

n1ck-guo committed 184 days ago

Verified 4e1b81f9

rm_old_vlm_cuda_ut (#678)

WeiweiZhang1 committed 185 days ago

Verified 1ead9fe9

rm low_cpu when loading the model (#676)

wenhuach21 committed 185 days ago

Verified 2fe91d77

restrict accelerate version to reduce ram usage (#673)

wenhuach21 committed 185 days ago

Verified 308e5e36

fix 3bits asym accuracy and calib dataset issues (#674)

WeiweiZhang1 committed 185 days ago

Verified bc14beaf

add trust remote code to gguf format load tokenizer (#675)

n1ck-guo committed 185 days ago

Verified 9461bf91

support to use lm_eval for vlm (#670)

n1ck-guo committed 185 days ago

Verified 16b8a262

fix cuda hanging issue during packing (#669)

WeiweiZhang1 committed 189 days ago

Verified 7863f69d

Older