intel/auto-round

Pull Requests Commits

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

pre-commit-ci[bot] committed 14 hours ago

6f5ed289

align act_max of experts for qwen3-vl and qwen3-next

xin3he committed 14 hours ago

ef05ec0a

open source delta loss (#1300)

wenhuach21 committed 1 day ago

Verified 587eadb8

fix ignore layers regression (#1302)

wenhuach21 committed 1 day ago

Verified 3352a830

Remove itrex format (#1301)

Kaihui-intel committed 1 day ago

Verified 46f88606

refine moe modellings to release orginal expert module's ram (#1265)

WeiweiZhang1 committed 2 days ago

Verified e383eea3

Update unit test script (#1290)

XuehaoSun committed 2 days ago

Verified 8e6f3370

Update version (#1282)

XuehaoSun committed 2 days ago

Verified 47c8aa4c

auto-round-kernel installation method (#1221)

chensuyue committed 6 days ago

Verified ba48dd77

[vllm-ext] allows setting flashinfer workspace (#1259)

yiliu30 committed 7 days ago

Verified 6e1ea9cb

add permission for workflow (#1190)

chensuyue committed 7 days ago

Verified 94db9057

Update HPU CI to gaudi v1.23.0 (#1271)

XuehaoSun committed 7 days ago

Verified 56750e8a

gguf fix bug of tensors not on same tensor and gguf:q2_k_mixed (#1263)

n1ck-guo committed 7 days ago

Verified 2c67712f

Fix critic bug causing GGUF to run on CPU (#1260)

wenhuach21 committed 9 days ago

Verified 1b7535a0

update supported scheme readme (#1257)

wenhuach21 committed 9 days ago

Verified 6a87a651

Update nightly release workflow (#1256)

chensuyue committed 9 days ago

Verified 583278d9

Add nightly release workflow (#1255)

chensuyue committed 9 days ago

Verified b4ab21d7

GGUF format add support for MoE models with non-linear expert layers. (#1244)

n1ck-guo committed 9 days ago

Verified c7d6aee6

fix cuda ut fail (#1237)

n1ck-guo committed 9 days ago

Verified 7a46aee8

Fix low cpu speed issue (#1251)

wenhuach21 committed 12 days ago

Verified 90db4445

WNA16 does not apply optimized RTN for moe layers by default (#1245)

wenhuach21 committed 12 days ago

Verified 9588bf90

Organize test directory structure with logical categorization (#1243)

Copilot committed 12 days ago

Verified 0f6dc76e

Update setting of disable opt rtn (#1249)

WeiweiZhang1 committed 13 days ago

Verified cc521cab

set disable_opt_rtn to optional bool and change default value to None (#1231)

WeiweiZhang1 committed 13 days ago

Verified bef28c9a

[HPU]Export lazy mode env explicit (#1239)

yiliu30 committed 13 days ago

Verified 001d0e39

[API Change]rename fp_layers to ignore_layers (#1233)

wenhuach21 committed 13 days ago

Verified 0ad224c9

Add FP8 dtype info to quant config (#1228)

yiliu30 committed 14 days ago

Verified afea815e

gguf add support for Mistral-Magistral-Devstral and Granite (#1226)

n1ck-guo committed 14 days ago

Verified 6f2f9b9e

[STEP 2] refactor format and export (#1192)

n1ck-guo committed 15 days ago

Verified c2449059

align input scale for moe experts for fp8 (#1216)

SwekeR-463 committed 15 days ago

Verified 41a83773

Older