intel/auto-round

Pull Requests Commits

Add envs.py for environment checks and update imports in test files (#1432)

xin3he committed 97 days ago

Verified 388e6c0b

reduce AutoGPTQ priority and add transformer requirement (#1467)

xin3he committed 97 days ago

Verified e7e1c78d

fix: qwen3-next NVFP4 quantization returns act_max incorrectly (#1470)

xin3he committed 97 days ago

Verified 229d1fef

Fix cuda ut workflow (#1472)

XuehaoSun committed 97 days ago

Verified b55ffb90

support glm5 (#1466)

wenhuach21 committed 99 days ago

Verified 62c9808d

Fix KeyError when GPU is missing from accelerate max_memory (#1457)

lvliang-intel committed 99 days ago

Verified b697e1d2

support multiple device evaluation for activation quantized model (#1394)

wenhuach21 committed 100 days ago

Verified bd4241df

Update compatibility test and version to 0.10.2 (#1463)

XuehaoSun committed 100 days ago

Verified 23c16c6c

Support load FP8 model on HPU (#1449)

yiliu30 committed 100 days ago

Verified aef6d3c4

fix: preserve classmethod descriptor in from_pretrained monkey patch (#1460)

yiliu30 committed 100 days ago

Verified 0d0f8367

Fix requirements packaging for source distribution (#1455)

timkpaine committed 111 days ago

Verified c0dc55fc

[Regression] fix FP8_STATIC loading (#1452)

xin3he committed 112 days ago

Verified 9855b4ba

Attach `act_max_hook` for FP8 model (#1447)

yiliu30 committed 112 days ago

Verified 81c8ee42

bump version (#1445)

chensuyue committed 112 days ago

Verified 0f087c91

fix bug of evaluate_accuracy (#1430)

xin3he committed 113 days ago

Verified 40253afe

update readme for Intel xpu usage (#1441)

chensuyue committed 113 days ago

Verified f16d57d7

update logic for compatibility with gptqmodel:exllamav2 backend (#1438)

xin3he committed 113 days ago

Verified 86231c44

Update auto-round-lib README.md (#1437)

chensuyue committed 113 days ago

Verified 232b1da9

Enable new cpu test pool (#1435)

chensuyue committed 113 days ago

Verified 0c7f4af3

Update diffusion README (#1439)

mengniwang95 committed 113 days ago

Verified 6e786c90

support GPTQ_FORMAT for "gptqmodel:exllamav2" backend (#1434)

xin3he committed 113 days ago

Verified 8a3d7937

refine global scale calculation to blockwise (#1421)

WeiweiZhang1 committed 114 days ago

Verified a554a107

fix CUDA CI (#1431)

xin3he committed 114 days ago

Verified d0f3cd0e

Update BackendInfos for AutoGPTQ based on transformer version and save g_idx for gptqmodel backend (#1429)

xin3he committed 114 days ago

Verified a1876107

Fix cuda model ut [glm4][Molmo] (#1428)

Kaihui-intel committed 114 days ago

Verified 16a18d7c

remove require_intel_extension_for_pytorch and fix TypeError: unhashable type: 'set' (#1425)

xin3he committed 114 days ago

Verified 319da040

fix bug in PR-1244 (#1422)

xin3he committed 115 days ago

Verified 81383e92

Refactor evaluation in tests to use evaluate_accuracy function (#1402)

xin3he committed 115 days ago

Verified 70a6bcc2

fix: update cleanup logic for quantized model path in tests (#1419)

xin3he committed 116 days ago

Verified ebb534a2

fix test_evaluation and AutoTuner new issue (#1417)

xin3he committed 116 days ago

Verified 520214a8

Newer Older