intel/auto-round

Pull Requests Commits

Support diffusion model AIDC-AI/Ovis-Image-7B quantization (#1616)

lvliang-intel committed 179 days ago

Verified 95a84d2b

fp8_block bug fix (#1693)

mengniwang95 committed 179 days ago

Verified 42101801

fix bug and simplify UT

xin3he committed 179 days ago

68d0cb7b

xin3he committed 179 days ago

687260db

remove high_gpu_mem_usage since no performacen benefit.

xin3he committed 179 days ago

83b9b4fe

xin3he committed 179 days ago

16991ea2

reduce flux CUDA CI tuning memory from 30G to 5GB (#1694)

xin3he committed 179 days ago

Verified 751216a4

remove useless code and update UT

xin3he committed 179 days ago

d3951f26

Fix hadamard transform weight dtype, using float32 as default and in-place transformed weight. (#1665)

lkk12014402 committed 179 days ago

Verified 43748dd9

update document

xin3he committed 179 days ago

58709e64

xin3he committed 180 days ago

4d9ad0e5

添加预编译模式匹配器以提高量化过程中的性能和可扩展性

xin3he committed 180 days ago

ff47a97a

remove useless gpu_concurrency

xin3he committed 180 days ago

97e03620

polished implementation

xin3he committed 180 days ago

177bf48b

Stabilize Qwen3-Omni MoE weight fidelity test for matching NaNs (#1681)

lvliang-intel committed 180 days ago

Verified 865f31b2

implement model free

xin3he committed 181 days ago

dc592e99

Enable NextStepDiffusion and support multi-device tuning for diffusion (#1640)

xin3he committed 181 days ago

Verified f81e20d6

Support MXINT4 scheme (#1666)

mengniwang95 committed 182 days ago

Verified c817d495

[mllm] support longcat_next (#1637)

xin3he committed 182 days ago

Verified 0821817b

Add compressed-tensors format export support for W4A16 and W8A16 (#1669)

thuang6 committed 182 days ago

Verified bbda38a6

Refactor module access to use PyTorch get_submodule / set_submodule (#1590)

scopophobic committed 185 days ago

Verified 58221b4a

Fix omni model test CI issue (#1667)

lvliang-intel committed 185 days ago

Verified bbda4046

Update unit test requirements (#1670)

XuehaoSun committed 186 days ago

Verified 440faaf1

Fix missing extra_config export for unsupported ignore_layers like `mlp.gate` (#1660)

lvliang-intel committed 186 days ago

Verified d2a62b89

import gptqmodel in confest to avoid error in transformers (#1668)

xin3he committed 187 days ago

Verified 106a3223

add gptqmodel 6.0 compatible change and MTP quantization skip warning. (#1663)

xin3he committed 188 days ago

Verified ac7aa288

fix autoscheme accuracy drop bug w/o low_gpu, add CI test (#1658)

WeiweiZhang1 committed 188 days ago

Verified c9a0d77b

Enhance performance test (#1610)

XuehaoSun committed 188 days ago

Verified 5b0d3faa

[pre-commit.ci] pre-commit autoupdate (#1661)

pre-commit-ci[bot] committed 188 days ago

Verified e4f94931

force auto-scheme low_gpu to True in CLI (#1653)

wenhuach21 committed 192 days ago

Verified 0f7dba7d

Newer Older