intel/auto-round

Pull Requests Commits

fix: remove transformers installation from unit test setup

chensuyue committed 3 days ago

89edcb1a

fix: update transformers version constraint in requirements file

chensuyue committed 3 days ago

354a0a03

fix: restrict transformers version to <5.15.0 in requirements files

chensuyue committed 3 days ago

7419549d

feat: add per-head fp8 kv attention calibration (#2044)

zeel2104 committed 5 days ago

Verified f17d9cd4

fix performance regression bug (#2136)

n1ck-guo committed 5 days ago

Verified f8ad63f4

Enhance unit tests with timeout handling and stability improvements (#2135)

XuehaoSun committed 6 days ago

Verified e4a02ccd

Add python 3.14t compatibility tests (#2110)

XuehaoSun committed 8 days ago

Verified 98f2cdb9

support one more mxfp4 variant (#2132)

wenhuach21 committed 8 days ago

Verified f706da67

Add llmc test back (#2131)

yiliu30 committed 8 days ago

Verified 1f6b2151

fix sparge on torch2.13 (#2130)

yiliu30 committed 9 days ago

Verified d8a30059

Refator UT and improve code coverage (#2057)

lvliang-intel committed 9 days ago

Verified 7a75218a

[fix] XPU crash due to circle call of nonzero() and index_select() (#2117)

xin3he committed 10 days ago

Verified 28003697

Enable torch.compile for selected tests and update related configurations (#2124)

xin3he committed 10 days ago

Verified a2f6b316

support fused MoE quantization and export for GGUF format (#2072)

n1ck-guo committed 10 days ago

Verified 1fca0c74

refactor scheme,layer_config and formats (#2013)

n1ck-guo committed 11 days ago

Verified 77e8ea59

Automatically select calibration datasets for code models (#2107)

changwangss committed 11 days ago

Verified f8de00c3

disable parallel for non low_gpu_mem_usage (#2116)

xin3he committed 12 days ago

Verified 417d87de

add opt_rtn support for mxfp4 and nvfp4 (#2100)

wenhuach21 committed 12 days ago

Verified 41cca72f

Fix: temporarily skip tests due to known issues with xfail annotations (#2113)

chensuyue committed 15 days ago

Verified 60b813cb

Enable AutoScheme parallel and disk stream by default (#2105)

xin3he committed 15 days ago

Verified 8a9a1d13

Fix: skip exporting fp8 attention q_max (#2103)

yiliu30 committed 15 days ago

Verified 2a267e72

fix MXFP model free auto-round format bug (#2106)

xin3he committed 15 days ago

Verified 378187c7

[Feature] Register_block_output to support GLM5.2 tuning (#2104)

xin3he committed 15 days ago

Verified d05fb638

feat: add OpenCodeInstruct calibration dataset (#2096)

changwangss committed 16 days ago

Verified ca9524ae

Enable torch.compile by default and update related documentation (#2070)

xin3he committed 16 days ago

Verified be862d37

Stream AutoScheme's per-layer sensitivity scoring block-by-block (#2063)

aquilarubra committed 16 days ago

Verified 73661edb

Fix dataset and batch_size not taking effect in CLI (#2102)

changwangss committed 16 days ago

Verified 212c8bda

Remove RoutedExperts since the bug is fixed. (#2098)

xin3he committed 17 days ago

Verified 8b0923c8

refine alg and support lfq (#2082)

wenhuach21 committed 17 days ago

Verified f1af646b

ark: add SpargeAttn support (#2027)

yiliu30 committed 17 days ago

Verified 5ce9733c

Older