intel/auto-round

Pull Requests Commits

Merge remote-tracking branch 'origin/main' into hengguo/reduce_cuda_ut

n1ck-guo committed 1 hour ago

38984587

Fix xpu CI retry issue (#2309)

chensuyue committed 1 hour ago

Verified fee969dd

Update CI/CD dockerfile (#2294)

XuehaoSun committed 2 hours ago

Verified 726becc3

reduce shard 0&1 runtime

n1ck-guo committed 3 hours ago

e36f9e36

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

wenhuach21 committed 2 days ago

Verified 7d8905ef

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

Copilot committed 2 days ago

Verified 21bab556

update awq detail data of fake eval (#2285)

WeiweiZhang1 committed 2 days ago

Verified 3942b57e

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

aquilarubra committed 2 days ago

Verified 63dbc8ba

fix gguf format mtp bug (#2297)

n1ck-guo committed 2 days ago

Verified 947ce102

fix reproducible seeds

jijiaz committed 2 days ago

55e4f787

modified repack fp16-bf16 route tests

jijiaz committed 2 days ago

be91de74

modified repack fp16-bf16 route tests

jijiaz committed 2 days ago

0474fab9

fix CUDA UT (#2284)

xin3he committed 2 days ago

Verified 039f027f

n1ck-guo committed 2 days ago

a0e7f7ce

Merge branch 'main' into hengguo/reduce_cuda_ut

n1ck-guo committed 3 days ago

Verified 00332e60

n1ck-guo committed 3 days ago

e1e2fc06

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

xesdiny committed 3 days ago

Verified c9017ff5

Merge branch 'hengguo/reduce_cuda_ut' of https://github.com/intel/auto-round into hengguo/reduce_cuda_ut

n1ck-guo committed 3 days ago

eb548659

n1ck-guo committed 3 days ago

6e7401c8

Merge branch 'main' into hengguo/reduce_cuda_ut

n1ck-guo committed 3 days ago

Verified c9d518d0

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

chensuyue committed 3 days ago

Verified b48c9db6

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

n1ck-guo committed 3 days ago

Verified 888b0f1e

n1ck-guo committed 3 days ago

664a9625

use tiny model and local json calibration

n1ck-guo committed 3 days ago

0fe82db7

use tiny model and move few e2e ut to weekly test

n1ck-guo committed 3 days ago

ff99bbe9

standardize desc of skip_cli

n1ck-guo committed 4 days ago

05a8359a

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

ramkrishs committed 4 days ago

Verified 81b4db35

Merge branch 'main' into hengguo/reduce_cuda_ut

chensuyue committed 4 days ago

Verified fd4a6958

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

avtc committed 4 days ago

Verified 5d2f8847

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

changwangss committed 4 days ago

Verified 1df1cef5

Older