intel/auto-round

Pull Requests Commits

enable low_cpu_mem_usage for mxfp/nvfp

Kaihui-intel committed 8 days ago

0ab96ea3

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

xin3he committed 13 days ago

Verified 47626ec5

fix CI for transformers==5.4.0 (#1627)

xin3he committed 14 days ago

Verified 5a38a394

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

lvliang-intel committed 14 days ago

Verified 11debae6

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

xin3he committed 15 days ago

Verified 1897e5ef

refine transform config api. (#1607)

lkk12014402 committed 15 days ago

Verified f07048b7

fix rtn bug (#1622)

wenhuach21 committed 15 days ago

Verified 891cf6ee

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

lvliang-intel committed 15 days ago

Verified b4d34c53

fix fp8 accuracy drop issue (#1611)

mengniwang95 committed 15 days ago

Verified b78453cd

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

WeiweiZhang1 committed 15 days ago

Verified 72d5c4dd

fix diffusion model tuning issue (#1617)

mengniwang95 committed 15 days ago

Verified e56369ca

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

XuehaoSun committed 15 days ago

Verified 1c27c468

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

chensuyue committed 16 days ago

Verified 51065799

support HPU memory dump in MemoryMonitor (#1605)

xin3he committed 16 days ago

Verified 6de3f6ae

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

xin3he committed 16 days ago

Verified 8f10ac54

support compressed-tensors refactor (#1595)

xin3he committed 17 days ago

Verified 19b52375

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

n1ck-guo committed 17 days ago

Verified a2aced88

fix CUDA UT for 0.12.0 (#1600)

xin3he committed 17 days ago

Verified 7da8ae09

Update ipex backend to be deprecated. (#1599)

xin3he committed 17 days ago

Verified 88824a12

fix contiguous issue (#1594)

xin3he committed 18 days ago

Verified 4689e93e

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

mengniwang95 committed 18 days ago

Verified d02b2ed4

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

Kaihui-intel committed 20 days ago

Verified 79fa1a9b

Support GLM-Image model quantizaiton (#1512)

lvliang-intel committed 20 days ago

Verified 62f92d1f

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

lvliang-intel committed 20 days ago

Verified 6bfe88fb

Add Performance Test (#1561)

XuehaoSun committed 21 days ago

Verified 29f02751

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

lkk12014402 committed 21 days ago

Verified a49550bc

Enable CUDA CI (#1473)

XuehaoSun committed 21 days ago

Verified 4f1a8dea

update awq backend infos, refine UT (#1581)

WeiweiZhang1 committed 21 days ago

Verified 06dfc838

disable torch compile for gguf with alg_ext (#1582)

wenhuach21 committed 21 days ago

Verified fcb0d2d1

[regression] fix copy missing tensors for llama4 (#1579)

xin3he committed 21 days ago

Verified 38c70355

Older