intel/auto-round

Pull Requests Commits

Add out-of-tree FP8 static quantization patches for Hunyuan-A13B-Instruct-FP8

yiliu30 committed 43 days ago

7428cfa1

[Feature] Enhance dataset preprocessing memory management and fix hash failure (#1621)

xin3he committed 44 days ago

Verified 7e77043e

Refine CI (#1644)

XuehaoSun committed 44 days ago

Verified 9f54985c

update xpu quantization cost (#1618)

xin3he committed 45 days ago

Verified e18a625c

Bump main branch version to v0.13.0 (#1639)

chensuyue committed 46 days ago

Verified 59c20f5e

Fix group_size issue (#1638)

mengniwang95 committed 46 days ago

Verified 7b0ff415

Update torch to 2.11.0 in CI (#1606)

XuehaoSun committed 46 days ago

Verified ae7661ba

skip overwriting the CUDA_VISIBLE_DEVICES and support transformers 5.4 (#1626)

xin3he committed 48 days ago

Verified 47626ec5

fix CI for transformers==5.4.0 (#1627)

xin3he committed 49 days ago

Verified 5a38a394

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

lvliang-intel committed 49 days ago

Verified 11debae6

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

xin3he committed 49 days ago

Verified 1897e5ef

refine transform config api. (#1607)

lkk12014402 committed 50 days ago

Verified f07048b7

fix rtn bug (#1622)

wenhuach21 committed 50 days ago

Verified 891cf6ee

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

lvliang-intel committed 50 days ago

Verified b4d34c53

fix fp8 accuracy drop issue (#1611)

mengniwang95 committed 50 days ago

Verified b78453cd

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

WeiweiZhang1 committed 50 days ago

Verified 72d5c4dd

fix diffusion model tuning issue (#1617)

mengniwang95 committed 50 days ago

Verified e56369ca

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

XuehaoSun committed 50 days ago

Verified 1c27c468

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

chensuyue committed 51 days ago

Verified 51065799

support HPU memory dump in MemoryMonitor (#1605)

xin3he committed 51 days ago

Verified 6de3f6ae

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

xin3he committed 51 days ago

Verified 8f10ac54

support compressed-tensors refactor (#1595)

xin3he committed 51 days ago

Verified 19b52375

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

n1ck-guo committed 52 days ago

Verified a2aced88

fix CUDA UT for 0.12.0 (#1600)

xin3he committed 52 days ago

Verified 7da8ae09

Update ipex backend to be deprecated. (#1599)

xin3he committed 52 days ago

Verified 88824a12

fix contiguous issue (#1594)

xin3he committed 53 days ago

Verified 4689e93e

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

mengniwang95 committed 53 days ago

Verified d02b2ed4

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

Kaihui-intel committed 54 days ago

Verified 79fa1a9b

Support GLM-Image model quantizaiton (#1512)

lvliang-intel committed 55 days ago

Verified 62f92d1f

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

lvliang-intel committed 55 days ago

Verified 6bfe88fb

Older