 Ben Li
|
90121e2521
|
change import from longcat_image to diffusers as upstream changes
|
2025-12-17 19:13:20 +08:00 |
|
|
|
8c2797a2bd
|
Sage attention speed boost (#6)
* Sage attention speed boost
Added attention_backend parameter to execute method and integrated SageAttention support.
* Document SageAttention backend for 2x inference speedup (#7)
* Initial plan
* Update README with SageAttention documentation
Co-authored-by: sooxt98 <13378059+sooxt98@users.noreply.github.com>
* Improve README clarity based on code review feedback
Co-authored-by: sooxt98 <13378059+sooxt98@users.noreply.github.com>
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: sooxt98 <13378059+sooxt98@users.noreply.github.com>
---------
Co-authored-by: Copilot <198982749+Copilot@users.noreply.github.com>
Co-authored-by: sooxt98 <13378059+sooxt98@users.noreply.github.com>
|
2025-12-10 11:39:02 +08:00 |
|
|
|
b6d27c34c7
|
Add CPU offload support for low VRAM GPUs (#1)
* Initial plan
* Implement low VRAM support with CPU offload option
Co-authored-by: sooxt98 <13378059+sooxt98@users.noreply.github.com>
* Address code review feedback - fix grammar and consistency
Co-authored-by: sooxt98 <13378059+sooxt98@users.noreply.github.com>
* Change default CPU offload setting to true
Updated the default value for 'enable_cpu_offload' to 'true' and adjusted the CPU offload logic for the transformer model loading process.
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: sooxt98 <13378059+sooxt98@users.noreply.github.com>
Co-authored-by: sooxt98 <sooxiaotong@gmail.com>
|
2025-12-08 20:22:35 +08:00 |
|
|
Sheldon Soo
|
a06e82b411
|
init
|
2025-12-06 20:45:57 +08:00 |
|