mirror of
https://github.com/comfyanonymous/ComfyUI.git
synced 2026-07-10 00:17:16 +08:00
Make chunking_mode a DynamicCombo on the Chunk SeedVR2 Latent node so frames_per_chunk is shown only when chunking_mode is manual. In auto mode the chunk size is predicted from free VRAM, so frames_per_chunk is irrelevant and is now hidden; temporal_overlap stays visible in both modes. Options are alphabetized (auto, manual). |
||
|---|---|---|
| .. | ||
| __init__.py | ||
| image_stitch_test.py | ||
| nodes_math_test.py | ||
| nodes_number_convert_test.py | ||
| test_seedvr2_conditioning.py | ||
| test_seedvr2_nodes.py | ||
| test_seedvr2_post_processing.py | ||
| test_seedvr2_temporal_chunk.py | ||