# ComfyUI v0.36.0 — retained record

> Retained reddit record from reddit:comfyui.

Canonical: https://highsignal.app/data/records/a8e1f507f2a61ce0

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:comfyui)  published 2026-09-19 14:37 ingested 2026-09-20 02:38

[original source ↗](https://www.reddit.com/r/comfyui/comments/1wknwkc/comfyui_v0360/)

## retained text



```
# Links

* [**Official changelog**](https://docs.comfy.org/changelog#v0-36-0)
* [**GitHub release**](https://github.com/Comfy-Org/ComfyUI/releases/tag/v0.36.0)

# New Open-Source Model Support

* [**FastVideo FastH3**](https://cloud.comfy.org/?template=video_fastvideo_fasth3_i2v): FastVideo 8-step distilled MiniMax H3 for text-to-video and image-to-video with native audio
* [**Marigold V2**](https://github.com/Comfy-Org/ComfyUI/pull/16232): Depth, surface normal, and albedo estimation from a single image
* [**YuE2**](https://github.com/Comfy-Org/ComfyUI/pull/16250): Music generation from style prompts and lyrics

# New Node Updates

* [**Video Concatenate**](https://github.com/Comfy-Org/ComfyUI/pull/16267): Join multiple video clips end to end
* [**Generic Loops**](https://github.com/Comfy-Org/ComfyUI/pull/16227): Loop over lists in the graph, carry heterogeneous items through every iteration, and declare loop boundaries in node schemas
* [**Linear color space**](https://github.com/Comfy-Org/ComfyUI/pull/16247): Added linear to ImageColorSpace
* [**Gemini V3 node**](https://github.com/Comfy-Org/ComfyUI/pull/16287): Added GeminiNodeV3 and deprecated the V2 node

# Partner Node Updates

* [**Tripo P2**](https://github.com/Comfy-Org/ComfyUI/pull/16320): Text, image, and multiview to 3D model generation with the P2 model
* [**Pruna P-Video-2**](https://github.com/Comfy-Org/ComfyUI/pull/16315): Text-to-video and image-to-video with 720p/1080p output and a draft mode
* [**Flux Video Edit**](https://github.com/Comfy-Org/ComfyUI/pull/16259): Edit a source clip from a text instruction: remove or replace objects, restyle footage, change on-screen text or dialogue
* [**Bria edit suite**](https://github.com/Comfy-Org/ComfyUI/pull/16264): New image edit nodes and the Video Eraser node

# Performance & Stability

* [**Auto fast-disk**](https://github.com/Comfy-Org/ComfyUI/pull/16333): Auto-detect and enable `--fast-disk` when the disk is fast, plus Aimdo 0.5.5
* [**MiniMax-H3 VAE**](https://github.com/Comfy-Org/ComfyUI/pull/16187): VAE optimizations and lower VAE memory usage
* [**Llama rope kernels**](https://github.com/Comfy-Org/ComfyUI/pull/16326): Use comfy-kitchen apply rope in llama-based models, with fast kernels on more models
* [**AMD Windows VA quota**](https://github.com/Comfy-Org/ComfyUI/pull/16199): Bumped the VA quota to 4TB
* [**YuE2 on AMD**](https://github.com/Comfy-Org/ComfyUI/pull/16293): Fixed AMD issues and added more controls to the generate ABC node
* [**YuE2 longer songs**](https://github.com/Comfy-Org/ComfyUI/pull/16292): Increased the max song duration to 900 seconds

# Bug Fixes

* [**Add Noise alpha**](https://github.com/Comfy-Org/ComfyUI/pull/15626): No longer adds noise to the alpha channel
* [**H3 Fun ControlNet**](https://github.com/Comfy-Org/ComfyUI/pull/16240): Fixed the H3 Fun ControlNet when the Comfy model compiler is active

# Workflow Templates

* [**Templates v0.11.62**](https://github.com/Comfy-Org/ComfyUI/pull/16350): Added FastH3 video, Marigold V2 estimation, and YuE2 music templates, and updated the Tripo, Gemini, and Ideogram workflows
```



document fetched 2026-09-20 00:37 · canonical https://reddit.com/r/comfyui/comments/1wknwkc/comfyui_v0360
