Commit Graph

Select branches

Hide Pull Requests

bindweave

causvid

longcat_avatar

longvie2

main

self-refine-video

syncammaster

vap

Mono Color

• 8321e19009 Fix single text encoder after previous change kijai 2025-08-01 14:45:29 +03:00
• 5406a72f62 Improve text encoder cache and add alternative node which leaves nothing to memory after encoding kijai 2025-08-01 12:36:19 +03:00
• 9edab74562 Default non_blocking to False kijai 2025-08-01 02:29:07 +03:00
• 7eebde487f typo kijai 2025-08-01 00:54:54 +03:00
• 945700d56e 5B fixes kijai 2025-07-31 22:41:56 +03:00
• 5bacc50088 Add safety check for context windows to avoid crash on tensor slicing kijai 2025-07-31 17:09:05 +03:00
• b60814fb6b Fix split prompting, get rid of diffusers on model class kijai 2025-07-31 15:11:47 +03:00
• 10b67de349 Fix loading TAEW from extra paths kijai 2025-07-31 13:47:59 +03:00
• 4cf2ffc979 Merge branch 'main' of https://github.com/kijai/ComfyUI-WanVideoWrapper kijai 2025-07-31 12:50:17 +03:00
• ef23a6213d return denoised samples as well for previewing unfinished latents kijai 2025-07-31 12:50:10 +03:00
• ee1e762a7f Update readme.md Jukka Seppänen 2025-07-31 12:37:16 +03:00
• 0cc4583653 flowedit fix kijai 2025-07-31 12:21:40 +03:00
• e545b18627 vid2vid fix kijai 2025-07-30 19:41:51 +03:00
• 4eabdcce78 wrong way around kijai 2025-07-30 18:45:19 +03:00
• 138b377d1d Fix loading fp8 models where everything is fp8... kijai 2025-07-30 17:34:56 +03:00
• 9a5d752ba1 Support custom sigmas on more schedulers kijai 2025-07-30 16:20:48 +03:00
• ec066008a8 Make start/end step work with vid2vid kijai 2025-07-30 15:59:22 +03:00
• 406643f8ee Create wanvideo_2_2_5B_I2V_example_WIP.json kijai 2025-07-30 00:46:46 +03:00
• 6ef3ee83ef Update nodes.py kijai 2025-07-30 00:40:42 +03:00
• 3f4e8433bd Add node to allow getting BasicScheduler sigmas kijai 2025-07-29 23:06:01 +03:00
• 1570287db1 Update nodes_model_loading.py kijai 2025-07-29 18:10:02 +03:00
• 15a57d22f7 fix slightly increased memory use kijai 2025-07-29 18:08:32 +03:00
• f835baa3d3 fix riflex kijai 2025-07-29 10:52:39 +03:00
• 01e452c0d1 Update wanvideo2_2_I2V_A14B_example_WIP.json kijai 2025-07-29 09:14:45 +03:00
• 234cffbf24 Fix comfy org scaled models giving black output kijai 2025-07-29 08:05:48 +03:00
• de11d52348 Update wanvideo2_2_I2V_A14B_example_WIP.json kijai 2025-07-29 02:46:01 +03:00
• 6875b99611 remove this for now kijai 2025-07-29 02:33:20 +03:00
• e4fb931878 Update wanvideo2_2_I2V_A14B_example_WIP.json kijai 2025-07-29 02:33:12 +03:00
• 5238f40fe3 Update nodes.py kijai 2025-07-29 02:20:20 +03:00
• a49936b645 Create wanvideo2_2_I2V_A14B_example_WIP.json kijai 2025-07-29 02:20:17 +03:00
• a3032e4363 clear lora patches first kijai 2025-07-29 01:45:25 +03:00
• 7e290c67bf Fix comfy progress bar when not doing full steps kijai 2025-07-28 23:16:06 +03:00
• fb7073259f detect 2.2 I2V kijai 2025-07-28 22:43:37 +03:00
• 498c6859fe Merge branch 'pr/883' kijai 2025-07-28 22:16:48 +03:00
• 08d4569794 Update nodes.py kijai 2025-07-28 21:42:35 +03:00
• d6e9af7090 2.2 stuff kijai 2025-07-28 21:38:40 +03:00
• fd285a7e7d implement tangential guidance kabachuha 2025-07-28 21:08:10 +03:00
• 998a69cc0a Update utils.py kijai 2025-07-28 10:46:46 +03:00
• 3659c3f6e6 Progress bar for text encoding kijai 2025-07-28 10:00:17 +03:00
• ee6e790d80 Update nodes.py kijai 2025-07-28 09:50:50 +03:00
• 5e725ff8fb Allow running T5 no cpu kijai 2025-07-28 09:39:36 +03:00
• 925ec31578 Merge pull request #876 from AustinMroz/main Jukka Seppänen 2025-07-27 19:33:11 +03:00
• 28da83c9b6 Fix animated latent preview compatibility Austin Mroz 2025-07-27 10:39:01 -05:00
• f61aae6158 minimax + context windows kijai 2025-07-27 17:10:41 +03:00
• f86d2b84a0 Update nodes.py kijai 2025-07-27 15:34:57 +03:00
• 3567202e75 Fix compile and support radial attn in the DF sampler kijai 2025-07-27 14:19:31 +03:00
• c0a4e8aa87 Update readme.md Jukka Seppänen 2025-07-27 01:43:32 +03:00
• bc68d35e5a Create wanvideo_1_3B_EchoShot_example.json kijai 2025-07-26 19:25:04 +03:00
• cb36e1013e Fix kijai 2025-07-26 19:01:58 +03:00
• df1476bf42 Support EchoShot kijai 2025-07-26 18:51:29 +03:00
• 9a1ab1c656 only show lora metadata if it actually exists kijai 2025-07-26 14:32:53 +03:00
• 5c36a4b111 minor memory tweaks kijai 2025-07-26 14:19:07 +03:00
• 3157e0389e Update nodes.py kijai 2025-07-26 11:50:24 +03:00
• 34c2cc23a9 Update nodes_model_loading.py kijai 2025-07-25 20:49:41 +03:00
• a35eb7dfa4 Basic sageattn3 support kijai 2025-07-25 20:48:34 +03:00
• 76750eec05 Allow loading aitoolkit/lycoris format kijai 2025-07-25 17:24:43 +03:00
• b89cf970e1 Fix GGUF + LoRA + VACE kijai 2025-07-25 16:44:43 +03:00
• 794a38cc7e Further GGUF+LoRA fixes kijai 2025-07-25 00:40:48 +03:00
• 73b7444baa remove print kijai 2025-07-24 23:54:30 +03:00
• ea36decf07 Fix GGUF + LoRA + torch.compile kijai 2025-07-24 23:52:18 +03:00
• b3820055d0 Fix GGUF with LoRAs and support GGUF witgh SetLoras -node kijai 2025-07-24 14:59:59 +03:00
• 35e637cbfd Allow merging LoRA to fp8 scaled models kijai 2025-07-24 11:51:36 +03:00
• d6425cab02 Update nodes_model_loading.py kijai 2025-07-24 00:08:39 +03:00
• 59cd0dc272 Update nodes.py kijai 2025-07-23 18:11:15 +03:00
• 838803d053 Update nodes_model_loading.py kijai 2025-07-23 17:58:28 +03:00
• bdce322e65 remove loras if SetLora bypassed/disconnected kijai 2025-07-23 17:51:59 +03:00
• 66226a8a1b why was this ever this way... kijai 2025-07-23 17:17:27 +03:00
• 3ed58330d9 better kijai 2025-07-23 16:56:33 +03:00
• 65ac9fa6d9 cache positive and negative separately, add cache to single encode node as well kijai 2025-07-23 16:33:11 +03:00
• 5e56e3649f fix control lora kijai 2025-07-23 16:20:30 +03:00
• 0212ad7a2d cache only the prompts kijai 2025-07-23 15:33:32 +03:00
• 1a1af02912 Allow using different init image for context windows beyond first one kijai 2025-07-23 12:51:57 +03:00
• 9d9b188b0b Update fp8_optimization.py kijai 2025-07-23 01:22:42 +03:00
• fa93f5b3c1 Add optional prompt disk caching kijai 2025-07-23 00:52:56 +03:00
• 729b6fdf7c support sparse sageattn2 for radial attn kijai 2025-07-23 00:27:14 +03:00
• 9c45f3a97d Update nodes_model_loading.py kijai 2025-07-23 00:10:14 +03:00
• eea896dace Update nodes_model_loading.py kijai 2025-07-22 18:25:03 +03:00
• 70fcdff3c5 Update nodes_model_loading.py kijai 2025-07-22 17:36:41 +03:00
• bfd6af8141 Update nodes_model_loading.py kijai 2025-07-22 15:41:32 +03:00
• 98fa84bb17 Update nodes_model_loading.py kijai 2025-07-22 13:58:43 +03:00
• 3988accdf3 Update nodes_model_loading.py kijai 2025-07-21 23:44:24 +03:00
• 9911b5e6f0 oops kijai 2025-07-21 22:49:05 +03:00
• 91515717d6 bump version kijai 2025-07-21 22:46:50 +03:00
• 296baa30ce Add WanVideoSetLoRAs kijai 2025-07-21 22:39:17 +03:00
• 41d8bd9ec9 fix for context windows kijai 2025-07-21 20:09:53 +03:00
• 4bf78e5316 Possibly reduce LoRA loading memory usage kijai 2025-07-21 15:25:07 +03:00
• 6ef9224dfe Update model.py kijai 2025-07-21 13:51:33 +03:00
• aeac12ed7a Only convert layers that actually have scaled weights... kijai 2025-07-21 01:57:27 +03:00
• 29ce253bb6 Support fp8_scaled models and allow running LoRAs unmerged on other models as well kijai 2025-07-21 01:37:30 +03:00
• 59add151eb Update attention.py kijai 2025-07-20 16:22:27 +03:00
• e5b125cd89 Allow setting any block as dense for Radial attention kijai 2025-07-20 16:21:37 +03:00
• 0dba876979 Correct radial attn mask size when using context windows kijai 2025-07-20 13:22:56 +03:00
• bb82d63453 Update nodes.py kijai 2025-07-20 13:08:40 +03:00
• 768a4b2d90 Fix radial attention cache not resetting on resolution change, add more robust supported dimension check kijai 2025-07-20 12:15:01 +03:00
• 9f75b7e051 Fix tiled_vae + endframe kijai 2025-07-20 02:33:24 +03:00
• 1f25bba463 Merge branch 'pr/828' kijai 2025-07-20 02:13:26 +03:00
• 47978275bb Update nodes.py kijai 2025-07-20 02:13:07 +03:00
• edd9b20691 Add res_multistep kijai 2025-07-20 01:41:17 +03:00
• d3aba90119 adding 64 block_size komikndr 2025-07-19 20:05:19 +07:00
• 0f6fe96626 Update __init__.py kijai 2025-07-18 23:01:03 +03:00