Commit Graph
119 Commits
Author SHA1 Message Date
kijai c4ed468967 Update nodes.py 2025-03-17 22:03:12 +02:00
kijai 7c8b4d09ae Update nodes.py 2025-03-17 20:18:46 +02:00
kijai 71737192fc Update nodes.py 2025-03-17 19:50:00 +02:00
kijai 6cd7f2296f Add Moebius latent shift looping
https://github.com/YisuiTT/Mobius/
2025-03-17 19:45:52 +02:00
James Banks f109df75c2 Update nodes.py to remove references to Hunyuan 2025-03-16 18:27:29 +00:00
kijai 173aff2367 fix I2V flowedit and maybe allow compile with vram management 2025-03-16 03:00:14 +02:00
kijai b18fafce2b Update nodes.py 2025-03-16 01:52:02 +02:00
kijai 81a08d0e59 Make non_blocking offloading optional for low RAM usecases 2025-03-16 01:48:45 +02:00
kijai f28d3f4225 somewhat working tiny vae previews 2025-03-15 23:12:38 +02:00
kijai 3aa573a74f Update nodes.py 2025-03-15 16:25:19 +02:00
kijai 58e5737032 Add option to use comfy native RoPE calculations
comfy rope doesn't use complex numbers and can be torch.compiled, then it's faster and uses less memory
2025-03-15 15:45:53 +02:00
kijai 66893bf39b taehv experiments 2025-03-15 13:06:29 +02:00
kijai 9576c39763 fix 2025-03-15 10:50:12 +02:00
kijai 1fbca61bf0 fix for VAE decoding in context loop mode 2025-03-15 01:25:51 +02:00
kijai 187eed3879 Update nodes.py 2025-03-15 00:50:02 +02:00
kijai ce79176646 support tiny vae 2025-03-15 00:48:55 +02:00
kijai 84a26d30f9 fix t2v context window 2025-03-14 12:08:57 +02:00
kijai 0dec61fde2 Update nodes.py 2025-03-13 23:48:02 +02:00
kijai af30809e11 control lora tweaks 2025-03-13 23:41:46 +02:00
kijai ff4bb8085e remove print 2025-03-13 18:17:46 +02:00
kijai 9f9ca8786d Update nodes.py 2025-03-13 17:29:42 +02:00
kijai 8ac0da07c6 Add SLG
Credits to AmericanPresidentJimmyCarter: https://github.com/deepbeepmeep/Wan2GP/pull/61
2025-03-13 17:29:15 +02:00
kijai 71fd6c7208 rope tweaks 2025-03-13 02:26:40 +02:00
kijai 36a0b0fb25 experimental context options 2025-03-12 23:44:08 +02:00
kijai 3a4b40d9ec Add tooltip to emphasize load_device -functionality 2025-03-12 23:18:34 +02:00
kijai e539839993 Update nodes.py 2025-03-12 23:16:07 +02:00
kijai c8677a211d Update nodes.py 2025-03-12 22:02:40 +02:00
kijai 83682769d4 Add low memory LoRA loading option 2025-03-12 21:55:35 +02:00
kijai b505075850 simplify LoRA loading further 2025-03-12 19:46:01 +02:00
kijai 0c780581f0 fix control lora 2025-03-12 19:16:33 +02:00
kijai 76cbbca6bd Update nodes.py 2025-03-12 18:50:31 +02:00
kijai 5c63ff5bce Faster RoPE 2025-03-12 18:36:28 +02:00
kijai 69e477ce7b Update nodes.py 2025-03-11 19:27:07 +02:00
kijai a4b1a4be7a Add optional offloading of the main model before text encoding 2025-03-11 09:57:46 +02:00
kijai d59f0a6732 fix 2025-03-10 21:45:47 +02:00
kijai 3b79887c8a Update nodes.py 2025-03-10 21:36:10 +02:00
kijai 570aaae86f Update nodes.py 2025-03-10 21:35:15 +02:00
kijaiandspacepxl ffaf1afe4e Support control-lora
https: //huggingface.co/spacepxl/Wan2.1-control-loras/blob/main/wan2.1-1.3b-control-lora-tile-v0.1_comfy.safetensors
Co-Authored-By: spacepxl <143970342+spacepxl@users.noreply.github.com>
2025-03-10 21:31:06 +02:00
kijai 9e83063dcf Make masking work for dpm and unipc too 2025-03-10 16:03:46 +02:00
kijai 4f2133f15f Actually way too slow to do this on cpu.... 2025-03-10 14:45:07 +02:00
kijai 2f085b8511 Simplify LoRA loading and actually load on cpu, add basic masking with differential diffusion, support finetrainers T2V LoRAs directly 2025-03-10 14:39:58 +02:00
kijai ca70c91586 Update nodes.py 2025-03-09 01:12:38 +02:00
kijai c6a1132d7f fix some tooltips 2025-03-09 01:10:18 +02:00
kijai 389bbd343c Update TeaCache docs 2025-03-08 12:13:52 +02:00
kijai 90a0f8c788 Allow compiling whole model too 2025-03-07 16:45:53 +02:00
kijai bc11af31e4 Update nodes.py 2025-03-06 02:07:00 +02:00
kijai 3cefbf0d70 small updates 2025-03-06 02:01:24 +02:00
kijai f4e706156f Update nodes.py 2025-03-05 22:07:36 +02:00
kijai 32465c2787 TeaCache tweaks for context windowing and FlowEdit 2025-03-05 21:02:34 +02:00
kijai de3e0f9ef2 update TeaCache 2025-03-05 13:35:18 +02:00