intel/auto-round

Pull Requests Commits

enable MiniMax-Text-01 bs!=1 quant

WeiweiZhang1 committed 275 days ago

a024ebb5

WeiweiZhang1 committed 288 days ago

a3f07782

fix gemma3 cache data issue

WeiweiZhang1 committed 288 days ago

6929eecd

add light ut, fixtypos (#483)

WeiweiZhang1 committed 288 days ago

Verified 7b5a2f1b

support xpu in tuning and inference (#481)

wenhuach21 committed 288 days ago

Verified 1603573a

add restrict for exporting act-quant models (#480)

n1ck-guo committed 292 days ago

Verified af565087

Temporarily close qxk api for new release (#478)

n1ck-guo committed 292 days ago

Verified d2cbed96

rm redundant line break (#475)

WeiweiZhang1 committed 300 days ago

Verified 050594f1

support packing immediately in new quantization api to save ram usage (#466)

wenhuach21 committed 300 days ago

Verified 8033dd0d

Increase unit test timeout from 120 to 240 minutes (#474)

XuehaoSun committed 300 days ago

Verified 34336d72

support to export gemma3 gguf format (#470)

n1ck-guo committed 300 days ago

Verified 43629a72

support for mistral3 (#472)

n1ck-guo committed 301 days ago

Verified 4e83d784

fix incorrect recipe data (#471)

WeiweiZhang1 committed 301 days ago

Verified 002ca16d

GGUF support step2: add naive Q2_KS and Q4_KS (#448)

n1ck-guo committed 301 days ago

Verified 12539ef5

upload_auto-round-light results (#454)

WeiweiZhang1 committed 302 days ago

Verified eec34cd6

support for gemma3 (#468)

n1ck-guo committed 306 days ago

Verified 451c029b

save bf16 to int4 scale and add HPU ut back (#467)

yiliu30 committed 306 days ago

Verified 5ed7e574

W4Afp8 export (#378)

wenhuach21 committed 309 days ago

Verified 7f6e5d40

fixl eval bug (#461)

n1ck-guo committed 310 days ago

Verified 5b95a8e6

20x for awq and 4x for gptq packing speedup (#459)

wenhuach21 committed 313 days ago

Verified c803ce9b

Support force loading into autoround Format (#453)

WeiweiZhang1 committed 313 days ago

Verified a79d1a36

awq exporting bugfix (#456)

WeiweiZhang1 committed 314 days ago

Verified 5ccdb0d1

update eval (#450)

n1ck-guo committed 314 days ago

Verified d51bd5e0

update readme (#455)

wenhuach21 committed 314 days ago

Verified 6bc03336

fix critic bug of mxfp4 (#451)

wenhuach21 committed 316 days ago

Verified abe69c54

step-1 support naive double quant in tuning (#442)

wenhuach21 committed 317 days ago

Verified 66adfb16

bump to v0.4.6 (#449)

XuehaoSun committed 322 days ago

Verified 1a9205a4

set torch compile to false by default (#447)

WeiweiZhang1 committed 327 days ago

Verified b6501629

Refine funcs (#446)

WeiweiZhang1 committed 329 days ago

Verified 3a143284

fix bug of evaluate user model (#444)

n1ck-guo committed 330 days ago

Verified 9aa88036

Older