Skip to content

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: use OpenS2V for diffusion calibration (#2368)
Wang, Chang committed 2 months ago
924b398
Honour the padding mask in SignRoundV2's plain loss (#2395)
Shubham Padkonde committed 2 months ago
0883217
Move the zero point to CPU for asymmetric layers too (#2394)
Shubham Padkonde committed 2 months ago
89c3dfe
Apply autocast only when mixed precision is enabled (#2393)
Shubham Padkonde committed 2 months ago
3f6c819
Cast float32 weights to the export dtype when saving (#2385)
Shubham Padkonde committed 2 months ago
0aa5e35
36079c4
Remove deprecated files (#2373)
chen, suyue committed 2 months ago
fd07787
remove bagel specific constraints (#2372)
Weiwei committed 2 months ago
4b3680b
3938e15
docs: clarify diffusion export format (#2367)
Wang, Chang committed 2 months ago
ea838e5
Add blockwise rotation. (#2087)
lkk committed 2 months ago
9b746ff
fix: save the config of diffusers models on export (#2119)
Nikita Davidchuk committed 2 months ago
1983b65