explicitly flag h1 as large model

Signed-off-by: Vladimir Mandic <mandic00@live.com>
This commit is contained in:
Vladimir Mandic
2025-04-14 10:25:31 -04:00
parent ac76138616
commit b63afc80fa
2 changed files with 5 additions and 4 deletions
+4 -3
View File
@@ -9,13 +9,14 @@
- add **UniPC FlowMatch** scheduler
- **HiDream** add HF gated access auth check
- clenup **CogView3** and **CogView4** model loader
- ui display reference models with subdued color
- xyz grid support bool
- add explicit offload after encode prompt
configure in *settings -> text encoder -> offload*
- networks: set which networks to skip when scanning civitai
in *settings -> networks -> network scan*
comma-separate list of regex patterns to skip
- ui display reference models with subdued color
- xyz grid support bool
- fix debug logging
- add explicit offload after encode prompt
## Update for 2025-04-12
+1 -1
View File
@@ -10,7 +10,7 @@ from modules.timer import process as process_timer
debug_move = log.trace if os.environ.get('SD_MOVE_DEBUG', None) is not None else lambda *args, **kwargs: None
should_offload = ['sc', 'sd3', 'f1', 'hunyuandit', 'auraflow', 'omnigen', 'cogview4']
should_offload = ['sc', 'sd3', 'f1', 'h1', 'hunyuandit', 'auraflow', 'omnigen', 'cogview4']
offload_hook_instance = None
balanced_offload_exclude = ['OmniGenPipeline', 'CogView4Pipeline']