intel/auto-round

Pull Requests Commits
Add Performance Test (#1561)
XuehaoSun committed 78 days ago
Verified 29f02751
support hadamard transform for mxfp4 with rtn or autoround method (#1515)
lkk12014402 committed 78 days ago
Verified a49550bc
Enable CUDA CI (#1473)
XuehaoSun committed 78 days ago
Verified 4f1a8dea
update awq backend infos, refine UT (#1581)
WeiweiZhang1 committed 78 days ago
Verified 06dfc838
disable torch compile for gguf with alg_ext (#1582)
wenhuach21 committed 78 days ago
Verified fcb0d2d1
[regression] fix copy missing tensors for llama4 (#1579)
xin3he committed 78 days ago
Verified 38c70355
Fix compatibility with latest compressed_tensors compressor refactor (#1576)
yiliu30 committed 78 days ago
Verified 49d2dbb0
add rtn quant func (#1564)
mengniwang95 committed 78 days ago
Verified 1ff8d028
fix deadlock issue when do quantization on hpu. (#1571)
lkk12014402 committed 78 days ago
Verified 093324c1
Update document for Qwen Omni quantized models (#1568)
lvliang-intel committed 78 days ago
Verified 92f0aaa3
Refine CPU RAM offload strategy (#1544)
lvliang-intel committed 79 days ago
Verified c37add63
add gptqmodel:awq backend support and keep autoawq WQLinear_GEMM for compatibility (#1545)
WeiweiZhang1 committed 79 days ago
Verified 84448965
Refine model save folder name (#1549)
Kaihui-intel committed 79 days ago
Verified 68d5568b
Fix reload issue for lm_head quantization and update CI to cover it (#1563)
xin3he committed 79 days ago
Verified 13fbffee
Freeze compressed-tensors & llmc version temporary (#1562)
chensuyue committed 79 days ago
Verified 4cc0cce4
temporary pin compressed-tensors version for llmcompressor (#1560)
chensuyue committed 80 days ago
Verified 3a8c26e6
Add Google-style docstrings to auto_round/data_type/ modules
Copilot committed 80 days ago
9446bf15
Add Google-style docstrings to auto_round/utils/ modules
Copilot committed 80 days ago
979fed3e
docs: add Google-style docstrings to utils/__init__.py and utils/distributed.py
Copilot committed 80 days ago
93c31335
docs: add Google-style docstrings to auto_round/data_type/ modules
Copilot committed 80 days ago
79ee1e72
docs: add Google-style docstrings to schemes, sign_sgd, envs, alg_ext
Copilot committed 80 days ago
a2b26c2a
Fix American English spelling in docstrings and clarify base class behavior
Copilot committed 80 days ago
4144989d
Add Google-style docstrings to calib_dataset.py and formats.py
Copilot committed 80 days ago
4655a684
Support Qwen3 and Qwen2.5 Omni model quantization (#1404)
lvliang-intel committed 80 days ago
Verified 91af35ff
docs: add Google-style docstrings to autoround.py and wrapper.py
Copilot committed 80 days ago
06e0226c
Initial plan
Copilot committed 80 days ago
61adaaf5
Fix MoE for llama/gpt-oss (#1557)
yiliu30 committed 80 days ago
Verified 555052e7
fix missing change for compress_layer_names and add UTs (#1556)
xin3he committed 80 days ago
Verified 6d17bbfb
Fix shard finalization issue to prevent skipping non-LLM tail layers (#1548)
lvliang-intel committed 81 days ago
Verified 81e79a14
Optimize mtp tensor process by compiling layer names once for all (#1554)
xin3he committed 81 days ago
Verified 1111bcd4
Terms Of Use Privacy Policy Refund Policy Impressum