# Changelog All notable changes to this project will be documented in this file. ## 26.11.0 - 2026-11-08 ### ⚠️ BREAKING CHANGES New versions are now using CalVer, update the image tag accordingly in your compose file ! ### 🚀 Features * **comfy**: Bump to v0.44.0 * * **release**: Show breaking change notes in changelog * **release**: Move to CalVer versioning (vYY.M.MICRO) ## 0.41.0 - 2026-10-27 ### 🚀 Features * **comfy**: Bump to v0.43.0 * ## 0.40.0 - 2026-10-21 ### 🚀 Features * **comfy**: Bump to v0.42.1 * * **extensions**: Remove `comfyui-vibevoice` as it is not compatible with transformer>=5 * **sageattention**: Uninstall, need to rework the prebuilt wheels * **torch**: Bump to v2.14.1 ### 🐛 Bug Fixes * **lora-manager**: Persist settings * Until now, the lora-manager extension had his settings being deleted on container restart. * The settings are located in the system user `__lora_manager` in the profiles folder. ## 0.39.0 - 2026-10-11 ### 🚀 Features * **comfy**: Bump to v0.39.0 * ## 0.38.0 - 2026-10-03 ### 🚀 Features * **comfy**: Bump to v0.38.0 * ### 🐛 Bug Fixes * **local-image-gallery**: Add a fix to the scrolling overflow issue when using Nodes 2.0 ## 0.37.0 - 2026-09-23 ### 🚀 Features * **comfy**: Bump to v0.37.0 * * **extensions**: Add RES4LYF extension * Adds [ClownsharkBatwing/RES4LYF](https://github.com/ClownsharkBatwing/RES4LYF), an advanced sampling node pack with 69+ selectable samplers, regional/temporal conditioning, and style transfer support. ## 0.36.0 - 2026-09-18 ### 🚀 Features * **comfy**: Bump to v0.36.0 * * **python**: Bump python to 3.13.15 ## 0.35.0 - 2026-09-14 ### 🚀 Features * **comfy**: Bump to v0.35.0 * ## 0.34.0 - 2026-08-29 ### 🚀 Features * **comfy**: Bump to v0.34.0 * ## 0.33.0 - 2026-08-25 ### 🚀 Features * **comfy**: Bump to v0.33.1 * * **core**: Add a patch support for extensions (#37) * Introduces a new feature that allows users to [apply patches to extensions](patches/README.md) in the ComfyUI project. The patch support enables users to modify and enhance the functionality of extensions without directly altering the original source code. * Thanks @algonixdev for the patch support implementation. * **deps**: Bump pytorch to 2.13.0 ## 0.32.0 - 2026-08-05 ### 🚀 Features * **comfy**: Bump to v0.30.0 * * **extensions**: Add new custom nodes * Pre-install [lbouaraba/comfyui-krea2edit](https://github.com/lbouaraba/comfyui-krea2edit) -- Instruction-based, identity-preserving image editing using Krea 2. * Pre-install [lquesada/ComfyUI-Inpaint-CropAndStitch](https://github.com/lquesada/ComfyUI-Inpaint-CropAndStitch) -- Crop before sampling and stitch back after sampling to speed up inpainting. * Pre-install [willmiao/ComfyUI-Lora-Manager](https://github.com/willmiao/ComfyUI-Lora-Manager) -- Organize, preview, and integrate LoRA models with metadata and workflow support. * Pre-install [yolain/ComfyUI-Easy-Use](https://github.com/yolain/ComfyUI-Easy-Use) -- Optimizations and integrations of commonly used nodes to make ComfyUI easier to use. ## 0.31.0 - 2026-07-01 ### 🚀 Features * **comfy**: Bump to v0.27.0 * ## 0.30.0 - 2026-06-29 ### 🚀 Features * **comfy**: Bump to v0.26.0 * ## 0.29.0 - 2026-06-20 ### 🚀 Features * **comfy**: Bump to v0.25.1 * * **comfy**: Bump to v0.25.0 * * **deps**: Bump `uv` to 0.11.23 * **docker**: Bump cuda to 13.0.3 * **python**: Bump to 3.13.14 * **sage-attention**: Recompile for torch 2.12 ### 🐛 Bug Fixes * **docs**: Set correct defaults for PUID / PGID environment variables ## 0.28.0 - 2026-06-04 ### 🚀 Features * **comfy**: Bump to v0.24.0 * * **manager**: Added full support from the manager * Comfyture is providing a set of extensions which are now being integrated within the manager UI. * The manager UI is using `uv` as the pip backend to install dependencies. * Dependencies from pre-installed extensions will not reset dependencies from extensions installed from the manager UI. ### 🐛 Bug Fixes * **docker**: Default user UID / GID is 1000 * **extensions**: `comfyui-rgthree` ➔ `rgthree-comfy` * Fix loading assets such as CSS / JS for this extension. ## 0.27.0 - 2026-05-27 ### 🚀 Features * **comfy**: Bump to v0.22.0 * ### 🐛 Bug Fixes * **compose**: `--normalvram` flag has been removed * **entrypoint**: Improve perms fix to only affect mismatched files ## 0.26.0 - 2026-05-18 ### 🚀 Features * **comfy**: Bump to v0.21.1 * ## 0.25.0 - 2026-05-10 ### 🚀 Features * **comfy**: Bump to v0.20.1 * ## 0.24.0 - 2026-04-15 ### 🚀 Features * **comfy**: Bump to v0.19.0 * ## 0.23.0 - 2026-04-11 ### 🚀 Features * **comfy**: Bump to v0.18.2 * ## 0.22.0 - 2026-03-14 ### 🚀 Features * **comfy**: Bump to v0.17.1 * ## 0.21.0 - 2026-03-13 ### 🚀 Features * **comfy**: Bump to v0.17.0 * * **entrypoint**: Enable smart memory management since dynamic vram introduction * [Dynamic Vram](https://github.com/Comfy-Org/ComfyUI/discussions/12699): the massive memory optimization is now enabled by default. * **extensions**: Add new ComfyUI Frame Interpolation extension * Added [ComfyUI-Frame-Interpolation](https://github.com/Fannovel16/ComfyUI-Frame-Interpolation) nodes. ## 0.20.0 - 2026-03-12 ### 🚀 Features * **dependencies**: Bump pytorch/cuda to 2.10+cu130 ## 0.19.0 - 2026-03-09 ### 🚀 Features * **agents**: Add guidelines for adding new ComfyUI extensions * **comfy**: Bump to v0.16.4 * * **nodes**: Add SeedVR2 video upscaler extension * Pre-install [numz/ComfyUI-SeedVR2_VideoUpscaler](https://github.com/numz/ComfyUI-SeedVR2_VideoUpscaler) -- Official SeedVR2 Video Upscaler for ComfyUI ### 🐛 Bug Fixes * **tasks**: Refactor update logics ## 0.18.0 - 2026-02-28 ### 🚀 Features * **comfy**: Bump to v0.15.1 * ## 0.17.0 - 2026-02-26 ### 🚀 Features * **comfy**: Bump to v0.15.0 * ## 0.16.0 - 2026-02-12 ### 🚀 Features * **comfy**: Bump to v0.13.0 * * **tasks**: Add `update` task to auto-update ComfyUI and dependencies ## 0.15.0 - 2026-02-05 ### 🚀 Features * **comfy**: Bump to v0.12.3 * * **nodes**: Add `comfyui-lanpaint` nodes * [comfyui-lanpaint](https://github.com/scraed/LanPaint) * Universally applicable inpainting ability for every model. * **nodes**: Add `comfyui-ltxvideo` nodes * [comfyui-ltxvideo](https://github.com/Lightricks/ComfyUI-LTXVideo) * A collection of powerful custom nodes that extend ComfyUI's capabilities for the LTX-2 video generation model. ### 🐛 Bug Fixes * **manager**: Disable manager UI * The system to lock dependencies enter in conflict with the manager. * I'm disabling it temporarily because at container start, the python venv is sync with what has been locked in `pylock.toml`. This will remove any dependencies installed via the manager. * **models**: Ensure folder `latent_upscale_models` is created ## 0.14.0 - 2026-01-22 ### 🚀 Features * **comfy**: Bump to v0.10.0 * ## 0.13.0 - 2026-01-18 ### 🚀 Features * **comfy**: Bump to v0.9.2 * ## 0.12.0 - 2026-01-13 ### 🚀 Features * **comfy**: Bump to v0.9.1 * * ## 0.11.0 - 2026-01-11 ### 🚀 Features * **comfy**: Bump to v0.8.2 * * **core**: Moving to new ComfyUI URL * [ComfyUI repo will move to Comfy Org account by Jan 6](https://blog.comfy.org/p/comfyui-repo-will-moved-to-comfy) * **nodes**: Add custom node `comfyui-detail-daemon` * [Jonseed/ComfyUI-Detail-Daemon](https://github.com/Jonseed/ComfyUI-Detail-Daemon): Adjust sigmas that generally enhance details, and possibly remove unwanted bokeh or background blurring. * **nodes**: Pre-install comfyui-rmbg * A custom node engineered for advanced image background removal and precise segmentation of objects, faces, clothing, and fashion elements. * See [1038lab/ComfyUI-RMBG](https://github.com/1038lab/ComfyUI-RMBG) for details * **nodes**: Pre-install comfyui-qwenvl * This custom node integrates the powerful Qwen-VL series of vision-language models including the latest Qwen3-VL and Qwen2.5-VL. * This is the version without GGUF support. * See [1038lab/ComfyUI-QwenVL](https://github.com/1038lab/ComfyUI-QwenVL) for details * **python**: Prune the `uv` cache after syncing dependencies * **python**: Bump to 3.13.11 ### 🐛 Bug Fixes * **deps**: Av==14.4.0 does not have pre-build wheel for python 3.13 * Use `av>=15.0.0,<16` which has. * **deps**: Sentencepiece==0.2.0 does not have pre-build wheel for python 3.13 * Use `sentencepiece>=0.2.1` which has. * **entrypoint**: Increase uv http timeout for downloading some packages (cuda) * **perms**: Changed permission test logic ## 0.10.0 - 2025-12-20 ### 🚀 Features * **comfy**: Bump to v0.5.1 * Announcement for [v0.5.0](https://github.com/comfyanonymous/ComfyUI/releases/tag/v0.5.0) * Patch release [v0.5.1](https://github.com/comfyanonymous/ComfyUI/releases/tag/v0.5.1) * **sage-attention**: Use pre-compiled wheels with cuda 13 support * **sharing**: Add `matrix-nio` dependency for sharing support * **startup**: Enable the new manager * See [Meet the New ComfyUI-Manager](https://blog.comfy.org/p/meet-the-new-comfyui-manager) * **startup**: Enable fast latent previews by default (latent2rgb) ## 0.9.0 - 2025-12-14 ### 🚀 Features * **comfyui**: Bump to 0.4.0 * **cuda**: Bump cuda to 13.0.2 * **flash-attention**: Bump flash-attention to 2.8.3 * **tests**: Do not use local volumes when testing ### 🐛 Bug Fixes * **python**: Bump python to 3.12.12 ## 0.8.0 - 2025-11-30 ### 🚀 Features * **comfyui**: Bump to v0.3.75 * **deps**: Add support for video thumbnails in `ComfyUI_Local_Media_Manager` node * **nodes**: New node for `DepthAnythingV3` support ### 🐛 Bug Fixes * **entrypoint**: Old name in header * **readme**: Absolute url for screenshot * **storage**: Ensure model folders `onnx`, `sams` and `vibevoice` are presents ## 0.7.0 - 2025-11-21 ### 🚀 Features * **comfyui**: Bump to v0.3.69 * **core**: The project has a name: comfyture * **docs**: Add a contributing guide * **docs**: Screenshot, extensions table, storage mappings explained * **models**: Ensure `insightface` model directory exists ### 🐛 Bug Fixes * **cd**: Reduce verbosity for cliff * **changelog**: Disable rule MD024 (duplicate headings) ## 0.6.0 - 2025-11-16 ### 🚀 Features * **core**: Allow to run as different user * **docker**: Prepare a compose file * **docs**: Add a nice readme :-) * **tasks**: Bump versions and changelog in pull requests ### 🐛 Bug Fixes * **changelog**: Do not show tag message * **ci**: Do not show changelog header in github release body * **release**: Tag message was not formatted correctly ## 0.5.0 - 2025-11-14 ### 🚀 Features * **ci**: Split ci and cd workflows ### 🐛 Bug Fixes * **release**: Tag is final * **release**: Set execution bit ## 0.4.2 - 2025-11-14 ### 🐛 Bug Fixes * **ci**: Wrong docker hub password secret ref ## 0.4.1 - 2025-11-14 ### 🐛 Bug Fixes * **ci**: Missing checkout step for release job ## 0.4.0 - 2025-11-14 ### 🚀 Features * **ci**: Make a unique workflow for pr validation with pre-release * **ci**: Add a release to docker hub * **extensions**: Split from entrypoint the management of extensions * Extensions are not anymore managed from the entrypoint. * This is handled from `extensions.sh`. You can install your own set of extensions and add requirements in `requirements.in`. The process will be improved later. ### 🐛 Bug Fixes * **ci**: Simplify docker build process ## 0.3.0 - 2025-11-07 ### 🚀 Features * **ci**: Push docker image to docker hub in bigbrozer/comfyui-stack * **ci**: Push image tasks ## 0.2.0 - 2025-11-07 ### 🚀 Features * **ci**: Add a ci to build and push docker image ## 0.1.0 - 2025-11-06 ### 🚀 Features * **comfyui**: Bump to 0.3.66 * **comfyui**: Bump to 0.3.64 * **comfyui**: Bump to 0.3.62 * **comfyui**: Bump to 0.3.60 * **comfyui**: Bump to v0.3.59 * **comfyui**: Bump to v0.3.57 * **comfyui**: Bump to 0.3.55 * **comfyui**: Bump to 0.3.53 + flash attention * **comfyui**: Bump to 0.3.49 * **comfyui**: Bump to 0.3.46 * **nodes**: Add comfyui-wanvideowrapper nodes * **nodes**: Add comfyui-segment-anything-2 nodes * **nodes**: Removed ltxvideo and wanblockswap nodes * **nodes**: Install `wanblockswap` nodes * **nodes**: Install `vibevoice` and `gguf` nodes * **nodes**: Install `comfyui-promptstylers` custom node * **release**: Prepare for public release ### 🐛 Bug Fixes * **nodes**: Downgrade to flash-attention 2.8.2 for wan video wrapper * **nodes**: Miss deps for `local-image-gallery` nodes ### ⚡ Performance * **comfyui**: Enable sage attention