intel/auto-round

Pull Requests Commits

Merge origin/main and address vLLM review follow-ups

Copilot committed 22 days ago

Verified c6d4034c

[ARK CI] Enhance auto-round-lib CI test (#1801)

chensuyue committed 22 days ago

Verified 5bbe39ba

[refactor] decouple calibaration code (#1765)

n1ck-guo committed 22 days ago

Verified 71fb3506

reduce XPU memory usage in CI (#1812)

xin3he committed 23 days ago

Verified bc545633

set default DYNAMO_CACHE_SIZE_LIMIT to 16 (#1786)

wenhuach21 committed 23 days ago

Verified cb9b0b76

Add Dockerfile with torch installed, enhance image build logic to check for changes (#1811)

chensuyue committed 23 days ago

Verified 694ad39a

Fix mixed-precision accuracy regression when AutoScheme runs with CPU offloading and Hadamard rotation enabled (#1753)

lvliang-intel committed 23 days ago

Verified 65ae39e7

Fix GGUF-K RTN routing and fake eval dtype normalization (#1807)

n1ck-guo committed 23 days ago

Verified 351ba32a

Enhance ark test workflow (#1805)

chensuyue committed 23 days ago

Verified 72d7fc8a

honor detected model type for use_vllm_loading, harden vLLM fallback for object inputs, and add vLLM utils unit tests

changwangss committed 23 days ago

30258de7

Potential fix for pull request finding

changwangss committed 24 days ago

Verified b4a86eb2

Potential fix for pull request finding

changwangss committed 24 days ago

Verified ba1c8353

[pre-commit.ci] auto fixes from pre-commit.com hooks

pre-commit-ci[bot] committed 24 days ago

2e9f3e37

feat(vllm): integrate use_vllm_loading into new AutoRound architecture with CLI/context/model routing and VllmMixin

changwangss committed 24 days ago

ebd1e409

Fix FP8 CT export metadata for KV cache and attention (#1752)

yiliu30 committed 24 days ago

Verified 24df42ea

fix bug of gguf alg ext (#1796)

n1ck-guo committed 24 days ago

Verified f64a0d5f

fix new arch bug for llmc (#1781)

n1ck-guo committed 24 days ago

Verified aa29325a

Integrate AutoRound Lib (#1723)

Zhenzhong1 committed 25 days ago

Verified 253117eb

support model_free WOQ quantization (#1699)

xin3he committed 25 days ago

Verified f0013f09

remove IPEX related code, doc, and test (#1787)

xin3he committed 25 days ago

Verified b2584b2d

add notes (#1795)

wenhuach21 committed 27 days ago

Verified 68df95b6

Fix incompatible weight names (#1759)

mengniwang95 committed 27 days ago

Verified 4c1eb8b1

fix amp (#1767)

wenhuach21 committed 28 days ago

Verified 6add4c27

fix amp (#1768)

wenhuach21 committed 28 days ago

Verified 8288c48f

fix accuracy regression and check it in CUDA CI (#1785)

xin3he committed 28 days ago

Verified 36767481

Optimize CUDA CI and Code Scan workflows (#1770)

XuehaoSun committed 28 days ago

Verified a46ad817

support gptqmodel 7.0.0 and fix bug in CI (#1772)

xin3he committed 28 days ago

Verified 871f6f3c

clean and fix for new arch (#1761)

n1ck-guo committed 36 days ago

Verified 1c615f1c

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

xin3he committed 36 days ago

Verified c859fabc

Fix hpu error (#1766)

n1ck-guo committed 36 days ago

Verified 244091f9

Older