Commit Graph

  • a4a956dbbd
    Add backend primitive nodes () Chenlei Hu 2025-03-21 01:47:18 -0400
  • 8b9ce4ed18
    Update frontend to 1.13 () Chenlei Hu 2025-03-21 00:17:36 -0400
  • 4932544a79
    Update frontend to 1.13 Chenlei Hu 2025-03-20 22:11:28 -0400
  • 4fbbca40d4 Add control after generate to int primitive Chenlei Hu 2025-03-20 15:22:06 -0400
  • e7875459d0 Add backend primitive nodes Chenlei Hu 2025-03-20 14:50:23 -0400
  • e562ca4728
    Merge branch 'comfyanonymous:master' into master Red 2025-03-20 04:25:30 -0700
  • 3872b43d4b A few fixes for the hunyuan3d models. comfyanonymous 2025-03-20 04:52:31 -0400
  • 32ca0805b7 Fix orientation of hunyuan 3d model. comfyanonymous 2025-03-19 19:55:24 -0400
  • 05b4dc868a
    Merge branch 'comfyanonymous:master' into master Red 2025-03-19 15:53:59 -0700
  • 11f1b41bab Initial Hunyuan3Dv2 implementation. comfyanonymous 2025-03-19 16:19:50 -0400
  • e4d69eb8df
    Update diffusers_load.py tongyu 2025-03-19 21:38:15 +0800
  • 6897a1d077 support pip comfyui-manager Dr.Lt.Data 2025-03-19 22:24:04 +0900
  • e5b6ac9b27 Close image after loading SiqiZhang19 2025-03-18 22:06:38 -0700
  • 4879b47648 Merge branch 'master' into worksplit-multigpu Jedrzej Kosinski 2025-03-18 22:19:32 -0500
  • 60b459bb4c Change SaveWEBM node's output key from images to video video_output Chenlei Hu 2025-03-18 17:12:02 -0400
  • b1270e56c7
    Merge branch 'master' into upgrade Muhammad Hassan 2025-03-18 20:27:35 +0500
  • 216a81ab8a fix missing comfyui-frontend-package & spandrel packages in nixpkgs x807x 2025-03-18 17:14:39 +0800
  • 3b19fc76e3 Allow disabling pe in flux code for some other models. comfyanonymous 2025-03-18 05:09:25 -0400
  • 17376d7597 specify architecture hassan00dev 2025-03-18 11:15:15 +0500
  • 762bd29942
    Merge branch 'comfyanonymous:master' into nix-support 807 2025-03-18 12:50:03 +0800
  • 2d8867de2e support output normal and lineart once Terry Jia 2025-03-17 21:29:31 -0400
  • dac0710c88 Add GitHub Action for Docker images Bernhard Frauendienst 2023-09-09 13:57:29 +0200
  • 33f01eb0df Add Dockerfile Bernhard Frauendienst 2023-09-07 21:43:55 +0200
  • 5ccec33c22 Merge branch 'worksplit-multigpu' of https://github.com/comfyanonymous/ComfyUI into worksplit-multigpu Jedrzej Kosinski 2025-03-17 14:27:39 -0500
  • 219d3cd0d0 Merge branch 'master' into worksplit-multigpu Jedrzej Kosinski 2025-03-17 14:26:35 -0500
  • 7929084115
    Merge branch 'comfyanonymous:master' into master Red 2025-03-17 14:01:35 -0400
  • 50614f1b79 Fix regression with clip vision. comfyanonymous 2025-03-17 13:56:11 -0400
  • 19edb4f8a2
    Update README.md Red 2025-03-17 12:53:32 -0400
  • ab866279c8
    Update README.md Red 2025-03-17 12:50:46 -0400
  • e4050bc809 Made changes to README RedsAnalysis 2025-03-17 12:49:32 -0400
  • 11a182ef10
    Update half_eps rockerBOO 2025-03-17 12:09:01 -0400
  • 6ee74beec4
    Merge branch 'comfyanonymous:master' into master Red 2025-03-17 07:55:05 -0400
  • 796b752e0a
    Merge pull request from RedsAnalysis/amd-support Red 2025-03-17 07:36:56 -0400
  • 0229028809 updated .sh to make the build more interactive and also worked out the docker file bugs RedsAnalysis 2025-03-17 07:27:09 -0400
  • a52ecc6ea5 changed dockerfile to include amd's gpu and pytorch ver. RedsAnalysis 2025-03-17 06:48:58 -0400
  • 6dc7b0bfe3 Add support for giant dinov2 image encoder. comfyanonymous 2025-03-17 05:53:54 -0400
  • abc1326566
    Merge branch 'comfyanonymous:master' into nix-support 807 2025-03-17 15:11:44 +0800
  • 8a50599bd8
    Fix RenormCFG node for batches rockerBOO 2025-03-17 01:03:13 -0400
  • e8e990d6b8 Cleanup code. comfyanonymous 2025-03-16 06:29:12 -0400
  • 2e24a15905
    Call unpatch_hooks at the start of ModelPatcher.partially_unload () Jedrzej Kosinski 2025-03-16 05:02:45 -0500
  • fd5297131f
    Guard the edge cases of noise term in er_sde () chaObserv 2025-03-16 18:02:25 +0800
  • ce6014c8cb Guard the edge cases of noise term in er_sde chaObserv 2025-03-16 14:48:43 +0800
  • 7cf6b5101c
    Merge branch 'master' into sa_solver chaObserv 2025-03-16 10:28:26 +0800
  • b9c927d351 Added ReadMe section for docker nvidia RedsAnalysis 2025-03-15 10:22:42 -0400
  • c4ba399475 Merge branch 'master' into worksplit-multigpu Jedrzej Kosinski 2025-03-15 09:12:09 -0500
  • d27a54b414
    Merge pull request from RedsAnalysis/comfymanager Red 2025-03-15 08:58:04 -0400
  • 6270f0073e Only call unpatch_hooks in partially_unload if lowvram is possible Jedrzej Kosinski 2025-03-15 07:57:19 -0500
  • 81d232e4b1 Changed the base image to cuda:12.6.3-cudnn-runtime-ubuntu24.04 and made sure everything works RedsAnalysis 2025-03-15 08:49:47 -0400
  • e03bd9e0e3 Call unpatch_hooks at the start of ModelPatcher.partially_unload Jedrzej Kosinski 2025-03-15 07:37:26 -0500
  • 55a1b09ddc Allow loading diffusion model files with the "Load Checkpoint" node. comfyanonymous 2025-03-15 08:27:49 -0400
  • 3c3988df45 Show a better error message if the VAE is invalid. comfyanonymous 2025-03-15 08:26:36 -0400
  • 13a79662f4 Final changes have been made need to update Readme RedsAnalysis 2025-03-15 07:23:16 -0400
  • 34b77380c2 Added ComfyUI-Manager support, made changes to .sh file to create volume for custom_nodes RedsAnalysis 2025-03-15 06:26:59 -0400
  • 01ceafa498 Given that gzip compression parameters already exist in cli args, revert the hardcoded code for this meimeilook 2025-03-15 17:45:35 +0800
  • b460dd62df
    Merge pull request from RedsAnalysis/dockerfile-branch Red 2025-03-15 04:57:51 -0400
  • bfdf01931c Add regular query workflow templates from custom_nodes MakkiShizu 2025-03-15 16:46:57 +0800
  • 974d0b1f1a Added Confyui manager to dockerfile, made changes to .sh file to recieve message from main.py, add place holder for venv volume mount RedsAnalysis 2025-03-15 04:27:18 -0400
  • 9239819dc1
    Use uv Inflation 2024-11-06 10:55:43 +0800
  • 1c6acbaf1f Updated docker-build.sh to include error handling for venv cp and in docker-compose.yml added image name tag RedsAnalysis 2025-03-15 02:42:10 -0400
  • 1ac7710b8d Update docker-build.sh and docker-compose.yml RedsAnalysis 2025-03-15 02:32:58 -0400
  • 4b2b24906a Put node_info caching in memory,and now 30x speedup when loading the page. meimeilook 2025-03-15 14:08:57 +0800
  • 7ebd8087ff
    hotfix fe () Christian Byrne 2025-03-14 22:38:10 -0700
  • 154fbbd681 Add Docker build script, Dockerfile, and Docker Compose setup for GPU compatibility RedsAnalysis 2025-03-15 01:22:16 -0400
  • cc8715c0e7
    Merge branch 'comfyanonymous:master' into nix-support 807 2025-03-15 13:04:24 +0800
  • 17fb1042ae add nix support with flake x807x 2025-03-15 13:03:16 +0800
  • ac0d92172c hotfix fe bymyself 2025-03-14 21:39:38 -0700
  • c624c29d66
    Update frontend to 1.12.9 () Chenlei Hu 2025-03-14 18:17:26 -0400
  • a2448fc527 Remove useless code. comfyanonymous 2025-03-14 18:10:37 -0400
  • 9d1de4e522
    Update requirements.txt Chenlei Hu 2025-03-14 16:10:22 -0400
  • a5c09997a2
    Update frontend to 1.12.9 Chenlei Hu 2025-03-14 14:57:39 -0400
  • 6a0daa79b6 Make the SkipLayerGuidanceDIT node work on WAN. comfyanonymous 2025-03-14 10:55:19 -0400
  • 9c98c6358b
    Tolerate missing @torch.library.custom_op () FeepingCreature 2025-03-14 14:51:26 +0100
  • be6030732d Tolerate missing @torch.library.custom_op This can happen on Pytorch versions older than 2.4. FeepingCreature 2025-03-14 12:48:08 +0100
  • f4208a28fd add args: npu-device Auc 2025-03-14 08:13:12 +0000
  • 715d345170 [feat]: add easy cache implementation for load_torch_file zhaoyingzhuo 2025-03-14 07:24:30 +0000
  • d40ac012bc 1.use getattr speedup 2x times in load node info when you opening browser at first time. 2.enable gzip in return response, make json files 10x smaller meimeilook 2025-03-14 15:39:21 +0800
  • 7aceb9f91c
    Add --use-flash-attention flag. () FeepingCreature 2025-03-14 08:22:41 +0100
  • 62a5b4ee88
    Update attention.py comfyanonymous 2025-03-14 03:21:16 -0400
  • cc928a786d Merge branch 'master' into worksplit-multigpu Jedrzej Kosinski 2025-03-13 20:59:11 -0500
  • eaba79602f Add uv. rh-uvtest Robin Huang 2025-03-13 16:28:12 -0700
  • f90ec0e6e2 Add --use-flash-attention flag. This is useful on AMD systems, as FA builds are still 10% faster than Pytorch cross-attention. FeepingCreature 2025-03-13 18:34:32 +0100
  • 35504e2f93 Fix. comfyanonymous 2025-03-13 15:03:18 -0400
  • 299436cfed Print mac version. comfyanonymous 2025-03-13 10:05:15 -0400
  • 52e566d2bc
    Add codeowner for comfy/comfy_types () Chenlei Hu 2025-03-12 17:30:00 -0400
  • 9b6cd9b874
    [NodeDef] Add documentation on multi_select input option () Chenlei Hu 2025-03-12 17:29:39 -0400
  • 3fc688aebd
    Ensure the extra_args in dpmpp sde series () chaObserv 2025-03-13 05:28:59 +0800
  • 4ab90a57c4 Add codeowner for comfy/comfy_types Chenlei Hu 2025-03-12 15:31:46 -0400
  • 4b1b1754cd [NodeDef] Add documentation on multi_select input option Chenlei Hu 2025-03-12 15:02:46 -0400
  • f4411250f3 Repeat frontend version warning at the end. comfyanonymous 2025-03-12 07:13:40 -0400
  • 1dc27b9b57 Ensure the extra_args in dpmpp sde series chaObserv 2025-03-12 18:51:05 +0800
  • d2a0fb6bb0
    Add unwrap widget value support () Chenlei Hu 2025-03-12 06:39:14 -0400
  • ad1102fc1d Modify LoadImage node to also traverse subdirectories of the input directory rewolf 2025-03-12 09:43:09 +0900
  • 01015bff16
    Add er_sde sampler () chaObserv 2025-03-12 14:42:37 +0800
  • 81025ce2c4 Quiet unzip woctordho 2025-03-12 11:55:34 +0800
  • 928540daf4 Obtain include and libs folders from nuget in stable-release.yml woctordho 2025-03-12 11:46:27 +0800
  • ded828f50b nit Chenlei Hu 2025-03-11 19:47:01 -0400
  • b949f971dc Add unwrap widget value support Chenlei Hu 2025-03-11 19:40:53 -0400
  • 2330754b0e Fix error saving some latents. comfyanonymous 2025-03-11 15:07:00 -0400
  • 5cd7530637
    Update nodes_upscale_model.py yash-711 2025-03-11 18:16:11 +0530
  • 73d8e2237f Add er_sde sampler chaObserv 2025-03-11 20:45:48 +0800