10 Commits
Author SHA1 Message Date
Fayche 6611efeba8 feat: Add offload directory creation for model dispatch in BagelModelLoader 2025-06-06 01:50:47 +08:00
Fayche d040497330 refactor: Clean up comments and improve code readability in BagelModelLoader 2025-06-06 01:19:11 +08:00
Fayche 76c1737bfe fix: Revise device_map 2025-06-06 00:49:50 +08:00
Fayche c4b4a4df4b feat: Add optimal memory calculation for GPU based on quantization mode 2025-06-05 14:59:04 +08:00
Fayche e6b8df01db feat: Enhance BAGEL Model Loader with Dynamic Quantization Support
- Added support for dynamic quantization modes (BF16, NF4, INT8) in BagelModelLoader.
- Updated INPUT_TYPES to include quantization options and advanced settings.
- Enhanced VALIDATE_INPUTS to validate quantization modes and check for bitsandbytes availability.
- Modified load_model to handle different quantization modes and improved memory management.
- Updated model loading logic to accommodate quantization configurations.
- Adjusted cfg_renorm_min default values in BagelTextToImage and BagelImageEdit classes.
- Added bitsandbytes as a requirement in requirements.txt.
2025-06-05 02:16:16 +08:00
WildAiandArt K 3e3081bc88 add comfyui progress bar (#23)
* add comfyui progress bar

* cleanup

---------

Co-authored-by: Art K <wildminder@users.noreply.github.com>
2025-06-03 16:44:15 +08:00
Faych Chen 777a359273 Merge pull request #21 from neverbiasu/df
feat: Refactor inferencer and improve df11 model loading logic
2025-05-30 12:22:06 +08:00
Fayche 292a1a7bbb fix: set cfg_renorm_min=0 by default 2025-05-28 22:29:52 +08:00
Fayche 70a0dddad5 fix: Update model download paths 2025-05-26 15:24:27 +08:00
Fayche e2a051d943 feat: Init node and requirements 2025-05-26 12:14:58 +08:00