intel/auto-round

Pull Requests Commits

Merge branch 'main' into lvl/offload_cleanup

lvliang-intel committed 28 days ago

Verified c57567d5

support model_type == "inkling_mm_model" (#2214)

xin3he committed 28 days ago

Verified de801f9f

fix: defer packing for multi-transformer diffusion (#2314)

changwangss committed 28 days ago

Verified b4dc0cc7

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

pre-commit-ci[bot] committed 28 days ago

dd4beea2

Merge branch 'main' into lvl/offload_cleanup

lvliang-intel committed 28 days ago

Verified e0b931f9

fix: update quantization configuration handling in FakeFormat and related tests (#2311)

xin3he committed 28 days ago

Verified 777c4ea7

Enhance offload cleanup handling for exception cases

lvliang-intel committed 28 days ago

a34e5437

Optimize diffusion calibration step sampling (#2265)

changwangss committed 29 days ago

Verified ce90add6

add sdxl svdquant (#2251)

changwangss committed 29 days ago

Verified 1c3a5518

fix: handle mantissa round overflow (#2302)

xin3he committed 29 days ago

Verified 9478fcf8

support 5,6,7 bits on cuda and add humming backend (#2281)

wenhuach21 committed 29 days ago

Verified 1f23576a

Fix xpu CI retry issue (#2309)

chensuyue committed 29 days ago

Verified fee969dd

Update CI/CD dockerfile (#2294)

XuehaoSun committed 29 days ago

Verified 726becc3

[risky]reduce ram usage of loading moe without monkey patch and fix qwen3-flash-next group_size issue (#2291)

wenhuach21 committed 31 days ago

Verified 7d8905ef

Optimize INT4 MoE prefill/decode (sym) XPU kernels with dedicated w4a16 tiling (#2111)

Copilot committed 31 days ago

Verified 21bab556

update awq detail data of fake eval (#2285)

WeiweiZhang1 committed 31 days ago

Verified 3942b57e

Allow non-MLLM calibration datasets when not quantizing vision (#2065)

aquilarubra committed 31 days ago

Verified 63dbc8ba

fix gguf format mtp bug (#2297)

n1ck-guo committed 31 days ago

Verified 947ce102

fix reproducible seeds

jijiaz committed 32 days ago

55e4f787

modified repack fp16-bf16 route tests

jijiaz committed 32 days ago

be91de74

modified repack fp16-bf16 route tests

jijiaz committed 32 days ago

0474fab9

fix CUDA UT (#2284)

xin3he committed 32 days ago

Verified 039f027f

feat(orchestrator): add reference_output to quantize_block to skip collect_reference (#2289)

xesdiny committed 32 days ago

Verified c9017ff5

Enhance test logging and retry mechanism for unit tests (#2274)

chensuyue committed 32 days ago

Verified b48c9db6

fix: skip activation fake-quant in FakeActQuantLinear when act quantization is not configured (#2286)

n1ck-guo committed 32 days ago

Verified 888b0f1e

fix: replace bare except clauses with except Exception (#1981)

ramkrishs committed 33 days ago

Verified 81b4db35

fix: asymmetric zero-point packing, format-scoped 8-bit asym, W5/W6/W7 llm_compressor presets (#2273)

avtc committed 33 days ago

Verified 5d2f8847

fix(diffusion): use real images for I2V calibration (#2272)

changwangss committed 34 days ago

Verified 1df1cef5

Update RunPod management to the V2 API. (#2271)

XuehaoSun committed 34 days ago

Verified 323e48d4

refactor algorithm cli (#2140)

n1ck-guo committed 34 days ago

Verified 1fd913d9

Older