intel/auto-round

Pull Requests Commits

Merge branch 'main' into wangchang/fix_oom

changwangss committed 170 days ago

Verified 2a716d47

clean and fix for new arch (#1761)

n1ck-guo committed 178 days ago

Verified 1c615f1c

[MTP]split gate_up_proj and fix accu gap in rtn quantization (#1758)

xin3he committed 178 days ago

Verified c859fabc

Fix hpu error (#1766)

n1ck-guo committed 178 days ago

Verified 244091f9

fix qdq inference oom issue.

changwangss committed 179 days ago

5b33a1b4

delete unreproduced results for now (#1760)

ZaneMark committed 179 days ago

Verified ed8eb671

Fix vllm CUDA CI (#1750)

XuehaoSun committed 180 days ago

Verified 5e2dbdbe

New architecture for auto_round (#1542)

n1ck-guo committed 180 days ago

Verified 91585c7d

add support for MiMo-V2-Flash (#1718)

n1ck-guo committed 181 days ago

Verified e62d29d1

add warnings for lm_head activation scale fallback (#1728)

n1ck-guo committed 181 days ago

Verified 99d49574

[Experimental]Add MLX format export support and AutoScheme for vlm support (#1732)

wenhuach21 committed 182 days ago

Verified 0def9d63

Reduce xpu memory usage with patch_xpu_sdpa_drop_causal_mask (#1716)

xin3he committed 183 days ago

Verified dc8a6a44

Enhance llmc CI on XPU (#1483)

chensuyue committed 183 days ago

Verified 19b3c485

add small zimage test and fix bug (#1734)

xin3he committed 184 days ago

Verified 46b29910

Remove threaded packing from exporters (#1719)

yiliu30 committed 184 days ago

Verified 85a61926

fix gemma3 gguf ut fail (#1735)

n1ck-guo committed 184 days ago

Verified 98db0b3c

Create model_support_request.yml (#1738)

xin3he committed 184 days ago

Verified 71f7946c

Update pull_request_template.md (#1727)

xin3he committed 184 days ago

Verified 3837fc38

skip quantizing mtp.fc since vLLM doesn't support (#1731)

xin3he committed 185 days ago

Verified b3ef561d

Revert "ignore mtp.fc for qwen3_5 due to vllm failure (#1710)" (#1730)

xin3he committed 185 days ago

Verified c2344554

Support new model Qwen/Qwen3.6-35B-A3B (#1705)

lvliang-intel committed 186 days ago

Verified 2dda035b

refine int4 doc (#1720)

wenhuach21 committed 186 days ago

Verified 5b6f7ea7

[research feature] Introduce INT4 support at the algorithm level (#1641)

wenhuach21 committed 186 days ago

Verified 11e91869

ignore mtp.fc for qwen3_5 due to vllm failure (#1710)

xin3he committed 186 days ago

Verified 8428b984

add support for gemma4 model (#1655)

n1ck-guo committed 187 days ago

Verified 55977e7b

Update gaudi-docker to v1.24.0 & fix CUDA UT (#1708)

XuehaoSun committed 187 days ago

Verified 8d784505

update mtp quant for special cases (#1691)

xin3he committed 188 days ago

Verified 89769e8a

rename scheme INT8_W8A8 to INT8 (#1687)

thuang6 committed 189 days ago

Verified 6a3bb90f

fix: add missing run_mllm entry point alias (#1695)

JGSphaela committed 191 days ago

Verified 693e4bd7

Add Claude skills for AutoRound (#1686)

lvliang-intel committed 191 days ago

Verified 4abe4067

Older