intel/auto-round

Pull Requests Commits

wenhuach21 committed 1 year ago

c1d5dafa

use torch.compile by default for PyTorch versions 2.6 and above (#295)

wenhuach21 committed 1 year ago

Verified c922f5b3

[Experimental Feature]support for common hf multimodel (#276)

n1ck-guo committed 1 year ago

Verified e6432125

fix bug of backend (#294)

wenhuach21 committed 1 year ago

Verified 4f228717

fix ipex tqdm mismatch issue (#293)

wenhuach21 committed 1 year ago

Verified 487abd6f

Add ipex support for intel cpu (#292)

wenhuach21 committed 1 year ago

Verified 168a1f69

Refine code (#291)

wenhuach21 committed 1 year ago

Verified f41094a9

update readme (#287)

wenhuach21 committed 1 year ago

Verified 8efff6f4

fix mx_fp issues (#286)

wenhuach21 committed 1 year ago

Verified 99cff1fb

avoid deterministic algorithm warning in inference (#285)

wenhuach21 committed 1 year ago

Verified ba5be40a

update readme for cpu inference

wenhuach21 committed 1 year ago

Verified 141c149f

update readme for v0.3.1 release (#283)

wenhuach21 committed 1 year ago

Verified a3592220

refine AuoRound format and support marlin repacking (#280)

wenhuach21 committed 1 year ago

Verified 68138e82

qwen2_bugfix, add adamround vision UT (#281)

WeiweiZhang1 committed 1 year ago

Verified 7cfff967

refine eval (#282)

wenhuach21 committed 1 year ago

Verified afa9e262

[Important update]set full range sym as the default (#278)

wenhuach21 committed 1 year ago

Verified 00122bc6

adamround bugfix, refine import (#275)

WeiweiZhang1 committed 1 year ago

Verified a633aa70

change to even rounding for mantissa of mx_fp (#277)

wenhuach21 committed 1 year ago

Verified 98a9c755

fix mutable default value (#272)

wenhuach21 committed 1 year ago

Verified 8bf63f39

enable llama3.2-vision model quantization (#269)

WeiweiZhang1 committed 1 year ago

Verified 6b99d10a

keep the dtype after qdq (#268)

wenhuach21 committed 1 year ago

Verified 3a70be84

remove g_idx in gptq format (#267)

wenhuach21 committed 1 year ago

Verified fdfd9711

Update readme for VLM support and integration (#266)

wenhuach21 committed 1 year ago

Verified af3db170

Add a warning for improper export formats. (#265)

wenhuach21 committed 1 year ago

Verified be32686b

Fix 3bit packing for auto-gptq format (#264)

wenhuach21 committed 1 year ago

Verified 6ee91a9f

better support quant_lm_head for larger models (#263)

wenhuach21 committed 1 year ago

Verified 82322ac9

refine autoawq exporting code (#261)

wenhuach21 committed 1 year ago

Verified 6539d506

update eval and fix example (#260)

n1ck-guo committed 1 year ago

Verified 7816eea5

enable_qwen2-vl_quantization (#248)

WeiweiZhang1 committed 1 year ago

Verified 3275df93

fix preci (#258)

n1ck-guo committed 1 year ago

Verified 200fcddc

Older