intel/auto-round

Pull Requests Commits

for test only, need to revert before merge

chensuyue committed 1 day ago

ca390382

Enhance unit test script and CI pipeline for AI failure analysis

chensuyue committed 1 day ago

3fb413c3

Fix packing format in quantization config and update variable assignment in tests (#1857)

xin3he committed 2 days ago

Verified 64c0a5bc

Fix special-model predefined ignore layer filtering (#1863)

lvliang-intel committed 2 days ago

Verified 930b4b90

Fix SDPA bug (#1862)

luoyu-intel committed 2 days ago

Verified c51d9fe7

fix: preserve FP8 llmc metadata for mxfp8 (#1861)

yiliu30 committed 3 days ago

Verified fdd9e305

Fix gemma4 crash issue during quantizing (#1860)

lvliang-intel committed 3 days ago

Verified 2bb9042e

Enhance CI workflow (#1858)

chensuyue committed 3 days ago

Verified 87c22224

refactor quarot/spinquant rotation with simplying code. (#1849)

lkk12014402 committed 4 days ago

Verified c73f36b5

fix bug of qwen and gguf export (#1846)

n1ck-guo committed 4 days ago

Verified 90ff3472

Fix Qwen Omni quantization model issue for long form audio generation (#1698)

lvliang-intel committed 4 days ago

Verified b4034073

update doc (#1856)

wenhuach21 committed 4 days ago

Verified f06f9a16

fix save_quantized log conflict (#1845)

WeiweiZhang1 committed 4 days ago

Verified 27761988

refine/fix gptq format (#1853)

wenhuach21 committed 4 days ago

Verified 09ca5615

dynamic map checkpoint naming based on model objective. (#1840)

xin3he committed 4 days ago

Verified 6c764875

Security: HTTP requests are performed without timeout safeguards (#1683)

tomaioo committed 5 days ago

Verified fbff4173

Sage fast sfm (#1843)

luoyu-intel committed 5 days ago

Verified d52c2a15

Fix gpt-j-6b RTN RuntimeError (#1848)

lvliang-intel committed 5 days ago

Verified 28f42389

Fix Gemma4 KeyError sliding_attention issue (#1839)

lvliang-intel committed 6 days ago

Verified 09bc7437

Fix some CI issues (#1842)

lvliang-intel committed 6 days ago

Verified fd495491

Support Exporting Block-Wise FP8 AR Format (#1798)

Zhenzhong1 committed 9 days ago

Verified bf1c81ce

support quarot/spinquant rotation before quantization (#1797)

lkk12014402 committed 10 days ago

Verified 064cf312

Reduce VRAM usage of quantizing VLM models (#1777)

lvliang-intel committed 10 days ago

Verified 64a5817d

add new accuracy test for VLM loading and fix bug (#1829)

xin3he committed 10 days ago

Verified a577f101

Awq algorithm (#1749)

WeiweiZhang1 committed 10 days ago

Verified e8ccff29

add mimo-audio, Qwen-TTS model backbone quantization (#1810)

WeiweiZhang1 committed 10 days ago

Verified 04e14693

add inc integration unit test (#1820)

xin3he committed 10 days ago

Verified 948b482c

Reduce RAM usage of quantizing VLM models and fix some issues of quantizing gemma4 (#1791)

lvliang-intel committed 10 days ago

Verified 9696884c

Update dependencies in CI and installation scripts for cuda compatibility (#1825)

XuehaoSun committed 11 days ago

Verified bd0a1f1e

remove new arch warning (#1832)

n1ck-guo committed 11 days ago

Verified 5ce6308a

Older