- 28 Sep, 2025 1 commit
-
-
gushiqiao authored
-
- 27 Sep, 2025 1 commit
-
-
gushiqiao authored
-
- 15 Sep, 2025 1 commit
-
-
Yang Yong(雍洋) authored
-
- 11 Sep, 2025 1 commit
-
-
sandy authored
-
- 25 Aug, 2025 1 commit
-
-
Yang Yong(雍洋) authored
-
- 18 Aug, 2025 2 commits
-
-
helloyongyang authored
-
gushiqiao authored
-
- 15 Aug, 2025 1 commit
-
-
helloyongyang authored
-
- 14 Aug, 2025 3 commits
-
-
helloyongyang authored
-
helloyongyang authored
-
helloyongyang authored
-
- 08 Aug, 2025 1 commit
-
-
helloyongyang authored
-
- 06 Aug, 2025 1 commit
-
-
gushiqiao authored
-
- 05 Aug, 2025 2 commits
- 04 Aug, 2025 1 commit
-
-
helloyongyang authored
-
- 30 Jul, 2025 1 commit
-
-
gushiqiao authored
-
- 29 Jul, 2025 1 commit
-
-
helloyongyang authored
-
- 22 Jul, 2025 1 commit
-
-
GoatWu authored
-
- 20 Jul, 2025 3 commits
-
-
helloyongyang authored
-
helloyongyang authored
-
helloyongyang authored
-
- 12 Jul, 2025 1 commit
-
-
GoatWu authored
-
- 10 Jul, 2025 1 commit
-
-
Zhuguanyu Wu authored
* support dynamic cfg for cfg_distill * reformat files
-
- 02 Jul, 2025 1 commit
-
-
gushiqiao authored
Enable 720p model inference on low-spec GPUs/CPUs and accelerate T5/CLIP quantized models with vLLM operators
-
- 01 Jul, 2025 1 commit
-
-
gushiqiao authored
* Fix * Fix * Fix --------- Co-authored-by:gushiqiao <gushiqiao@sensetime.com>
-
- 16 Jun, 2025 1 commit
-
-
gushiqiao authored
-
- 22 May, 2025 1 commit
-
-
root authored
-
- 13 May, 2025 1 commit
-
-
wangshankun authored
-
- 29 Apr, 2025 1 commit
-
-
wangshankun authored
-
- 23 Apr, 2025 1 commit
-
-
helloyongyang authored
-
- 20 Apr, 2025 2 commits
-
-
helloyongyang authored
-
helloyongyang authored
-
- 08 Apr, 2025 3 commits
-
-
Dongz authored
* [minor]: optimize dockerfile for fewer layer * [feature]: add pre-commit lint, update readme for contribution guidance * [minor]: fix run shell privileges * [auto]: first lint without rule F, fix rule E * [minor]: fix docker file error
-
TorynCurtis authored
* 修改了main.py, t5的model, wan的model、三个weights文件和三个infer文件, 并且在common的conv3d算子中注册新算子 * 修改了Conv3dWeightForceBF16算子，更新了wan的pre_weights中对此算子的使用 * 修复了import中的bug * 修复了WanPreWeights, WanTransformerWeights没有self.config的bug * 修复了WanPreWeights, WanTransformerWeights没有self.config的bug * 修复了config的bug，目前在使用cpu_offload的时候，vae阶段有tensor不在同一device的bug * 修复了vae阶段迁移的bug * 修复了scale在mean和inv_std迁移后仍需重新赋值的bug
-
helloyongyang authored
-
