PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度
Commits on main
feat: add workflow to close stale issues and pull requests (#1612)
5106579 chen, suyue committed 8 months ago
fix: refactor and fix Omni model tests to improve CI stability (#1598)
8f10ac5 Xin He committed 8 months ago
fix packing nvfp/mxfp `max_wokers` & extend xpu ut (#1555)
79fa1a9 Tang Kaihui committed 8 months ago
support hadamard transform for mxfp4 with rtn or autoround method (#1515)
a49550b lkk committed 8 months ago
Fix compatibility with latest compressed_tensors compressor refactor (#1576)
49d2dbb Yi Liu committed 8 months ago
add gptqmodel:awq backend support and keep autoawq WQLinear_GEMM for compatibility (#1545)
8444896 Weiwei committed 8 months ago
Fix reload issue for lm_head quantization and update CI to cover it (#1563)
13fbffe Xin He committed 8 months ago
Freeze compressed-tensors & llmc version temporary (#1562)
4cc0cce chen, suyue committed 8 months ago
temporary pin compressed-tensors version for llmcompressor (#1560)
3a8c26e chen, suyue committed 8 months ago