intel/auto-round

Pull Requests Commits

Zhenzhong1 committed 70 days ago

fc811cc1

fix: warm up Triton CUDA driver before mocking device capability (#2253)

XuehaoSun committed 71 days ago

Verified a97b91a2

bump release version (#2257)

chensuyue committed 72 days ago

Verified dccf193d

disable torch compile in rtn/opt-rtn/iters<10 and fix glm-5.3 tuning exception (#2246)

wenhuach21 committed 74 days ago

Verified 4dbbd15d

[Bugfix] Fix Sglang e2e UT issue (#2250)

lvliang-intel committed 74 days ago

Verified d77c9eaa

fix argument mismatch (#2248)

xin3he committed 74 days ago

Verified 53f3802a

fix: preserve diffusion mixed dtypes during loading (#2232)

changwangss committed 74 days ago

Verified 7585f13c

fix: remove stale quantizer access in dual-transformer diffusion (#2236)

changwangss committed 74 days ago

Verified 68f8aa32

Update troubleshooting tips for quantization issues in README_CN.md (#2244)

XuehaoSun committed 75 days ago

Verified 7eb007ad

test: skip HPU FP8 test with transformers 5.16.x (#2247)

changwangss committed 75 days ago

Verified c3a74f30

increase precision of loss logging in quantization tuning for debugging (#2229)

xin3he committed 75 days ago

Verified 385a48c9

test: skip FP8 tests with transformers 5.16.x (#2241)

changwangss committed 75 days ago

Verified b4628f15

Fix CUDA UT issue (#2238)

xin3he committed 75 days ago

Verified 9e030fd1

Fix Cuda UT coverage collect (#2234)

chensuyue committed 76 days ago

Verified e5693be8

refactor uts, remove duplicate uts and build common uts for all devices to increase coverage (#2157)

n1ck-guo committed 76 days ago

Verified 6410bc20

fix: AutoScheme parallel scoring OOMs (MoE memory, VLM gate) and disk-stream tuning crash (#2220)

avtc committed 76 days ago

Verified 7da9bea8

Fix FP8 quantization config for native fp8 export (#2228)

Zhenzhong1 committed 76 days ago

Verified 017adc89

add rotation ut for xpu. (#2094)

lkk12014402 committed 76 days ago

Verified 60285730

add enable_torch_compile mark to SGLang and vLLM throughput tests (#2226)

xin3he committed 77 days ago

Verified 4b7cc7e8

Update alg combination news (#2227)

wenhuach21 committed 77 days ago

Verified bd9a13ce

update auto_scheme_acc doc (#2000)

n1ck-guo committed 77 days ago

Verified 5683ea15

add awq torch_compile fix, disable_opt_rtn by default for awq (#2208)

WeiweiZhang1 committed 77 days ago

Verified 611d4402

disable timeout in pytest until found a more stable method (#2222)

chensuyue committed 77 days ago

Verified 96ce4480

Preserve diffusion component dtypes during quantization (#2205)

changwangss committed 78 days ago

Verified d6ead479

refactor: optimize expert selection and output handling in moe forward (#2209)

xin3he committed 80 days ago

Verified b9f3d007

add workaround for llmc issue and add unit tests for model_free_utils helpers (#2207)

xin3he committed 80 days ago

Verified a2c27b01

Fix CT export in mixed‑precision mode (#2128)

1607662794 committed 80 days ago

Verified d034df53

fix: layer_config for lm_head takes effect in model-free mode (#2182)

Copilot committed 81 days ago

Verified 1d050463

feat: add support for FP8 model loading and ragged-tile dequantization fallback (#2158)

xin3he committed 81 days ago

Verified 40635d02

Reduce SYCL-TLA Compiler Memory (#2193)

luoyu-intel committed 81 days ago

Verified ff2fd25c

Older