intel/auto-round

Pull Requests Commits

Bump to v0.16.1 (#2517)

XuehaoSun committed 26 days ago

Verified cbdec9f3

fix: use spawn start method for vLLM evaluation to avoid CUDA re-initialization error (#2421)

reginaldalfret committed 26 days ago

Verified 234ac7af

fix: preserve calibration dataset on memory errors (#2433)

Shubham-Padkonde committed 26 days ago

Verified fde1f580

feat: support LLMC/ModelOpt/Quark format NVFP4 -> MXFP4 conversion (#2492)

xin3he committed 27 days ago

Verified 7222a20e

Support Nemotron-H Omni wrappers (Nemotron 3.5 Super VL, Bolt 74B-A9B) in plain AutoRound calls (#2512)

SergiioB committed 29 days ago

Verified ec10a649

feat: implement Kimi remote MoE with grad-aware inference and add uni… (#2496)

xin3he committed 29 days ago

Verified 88dc729d

remove leaderboard entry for now as it's still not very stable (#2516)

wenhuach21 committed 29 days ago

Verified 650981e0

docs(examples): align recipes with recommended W2 and batch size settings (#2499) (#2511)

vishvaRam committed 30 days ago

Verified a0134a4c

change slogan (#2509)

wenhuach21 committed 30 days ago

Verified 08d7085e

[Experimental]Add generic HF dataset support and two more datasets (#2363)

wenhuach21 committed 30 days ago

Verified e31b7728

fix: adjust tolerance levels for performance metrics comparison (#2505)

XuehaoSun committed 30 days ago

Verified 2c377483

fix: quantize all diffusion transformers in RTN routes (#2497)

changwangss committed 30 days ago

Verified 9645350f

Add clear entry for leaderboard (#2500)

wenhuach21 committed 31 days ago

Verified 523251eb

fix gemma4 12b vllm issue (#2491)

wenhuach21 committed 31 days ago

Verified 7ef18f8c

feat(examples): add batch 2 quantization recipes for Muse-Glimmer, NuExtract3, and Qwen3.5 (#2498)

vishvaRam committed 31 days ago

Verified da52d6da

fix: enable imatrix calibration for MX and NV optimized RTN (#2494)

changwangss committed 31 days ago

Verified 94aa6d40

fix: enhance attention module with query observation and legacy support (#2466)

xin3he committed 31 days ago

Verified 376fdc0f

Add quantization recipe notebooks for Qwen, OCR, and edge models (#2488)

vishvaRam committed 32 days ago

Verified 504aea5c

Don't rewrite packed modules as float weights when resuming (#2386)

Shubham-Padkonde committed 33 days ago

Verified 4663212a

fix: update field extraction for ServerArgs in _run_sglang_worker (#2474)

xin3he committed 33 days ago

Verified a0651215

fix: update dependencies to include timm in unit test environment (#2473)

XuehaoSun committed 34 days ago

Verified 421ae1af

fix: skip layers the AWQ GEMM kernel cannot serve in AWQ export (#2456)

losewayy committed 35 days ago

Verified 7a660e16

fix: remove duplicate dataset aliases (#2470)

changwangss committed 35 days ago

Verified 0027ecff

bump version for auto-round-lib (#2458)

chensuyue committed 35 days ago

Verified 57230c2b

fix: enhance model evaluation preparation for non-task-based evaluation (#2468)

xin3he committed 35 days ago

Verified 1f421d6d

fix torch compile (#2459)

wenhuach21 committed 35 days ago

Verified 799ebee7

fix: keep model obj evaluation for regular workflow (#2454)

xin3he committed 36 days ago

Verified ea35ee11

fix: match layer names ending with a digit in ignore_layers (#2435)

kwy404 committed 36 days ago

Verified 5aef837e

Add requests into main requirements.txt (#2441)

chensuyue committed 36 days ago

Verified cba7a4f4

fix: forward pipeline call kwargs through diffusion entry (#2448)

changwangss committed 37 days ago

Verified 63895ebf

Newer Older