intel/auto-round

Pull Requests Commits

fix fp8 expert patch

xin3he committed 59 days ago

6fbca356

fix gguf buf of transformers 5.4 (#1632)

n1ck-guo committed 59 days ago

Verified 06d1c110

remove bias since qwen don' t have it

xin3he committed 59 days ago

7347a9af

Merge branch 'main' into xinhe/3-27

xin3he committed 59 days ago

Verified a25f75c9

fix CI for transformers==5.4.0 (#1627)

xin3he committed 59 days ago

Verified 5a38a394

Apply suggestion from @xin3he

xin3he committed 59 days ago

Verified 88e5f40f

skip overwriting the CUDA_VISIBLE_DEVICES

xin3he committed 59 days ago

f01400b4

Fix AutoScheme low memory flag propagation from CLI (#1596)

lvliang-intel committed 59 days ago

Verified 11debae6

Allow disabling copy_missing_tensors_from_source via env var, default enabled (#1623)

xin3he committed 59 days ago

Verified 1897e5ef

refine transform config api. (#1607)

lkk12014402 committed 60 days ago

Verified f07048b7

fix rtn bug (#1622)

wenhuach21 committed 60 days ago

Verified 891cf6ee

Fix clean offload clearing of non-checkpoint tensors (#1583)

lvliang-intel committed 60 days ago

Verified b4d34c53

fix fp8 accuracy drop issue (#1611)

mengniwang95 committed 60 days ago

Verified b78453cd

revert awq dtype change for vllm inference limitation (#1613)

WeiweiZhang1 committed 60 days ago

Verified 72d5c4dd

fix diffusion model tuning issue (#1617)

mengniwang95 committed 60 days ago

Verified e56369ca

Fix LLMC CI & Update CPU CI to transformers 5.x (#1604)

XuehaoSun committed 60 days ago

Verified 1c27c468

feat: add workflow to close stale issues and pull requests (#1612)

chensuyue committed 61 days ago

Verified 51065799

support HPU memory dump in MemoryMonitor (#1605)

xin3he committed 61 days ago

Verified 6de3f6ae

fix: refactor and fix Omni model tests to improve CI stability (#1598)

xin3he committed 61 days ago

Verified 8f10ac54

support compressed-tensors refactor (#1595)

xin3he committed 61 days ago

Verified 19b52375

fix gguf bug for qwen3.5 moe and some small models (#1575)

n1ck-guo committed 62 days ago

Verified a2aced88

fix CUDA UT for 0.12.0 (#1600)

xin3he committed 62 days ago

Verified 7da8ae09

Update ipex backend to be deprecated. (#1599)

xin3he committed 62 days ago

Verified 88824a12

fix contiguous issue (#1594)

xin3he committed 62 days ago

Verified 4689e93e

Revert "add rtn quant func (#1564)" (#1593)

mengniwang95 committed 63 days ago

Verified d02b2ed4

fix packing nvfp/mxfp `max_wokers` & extend xpu ut (#1555)

Kaihui-intel committed 64 days ago

Verified 79fa1a9b

Support GLM-Image model quantizaiton (#1512)

lvliang-intel committed 65 days ago

Verified 62f92d1f

Fix CI test FP8 static backend matching failure (#1589)

lvliang-intel committed 65 days ago

Verified 6bfe88fb

Add Performance Test (#1561)

XuehaoSun committed 65 days ago

Verified 29f02751

support hadamard transform for mxfp4 with rtn or autoround method (#1515)

lkk12014402 committed 65 days ago

Verified a49550bc

Older