ComfyUI v0.43 adds MOSS-Image layout control and edit LoRAs, ACEScg EXR output and Kling 4.0 nodes
ComfyUI v0.43.0, released on 27 October, is the first stable build with native support for MOSS-Image color and depth control and for MOSS-Image edit LoRAs, which until now needed a nightly build. It also adds ACEScg colour space to EXR saving, new Pad Image and Branch nodes, and partner nodes for Kling 4.0, PixVerse V6-Max, FLUX 3 Max and Ideogram Vector.
What it does
For local models, the release brings the MOSS-Image Color/Depth Control node, support for MOSS-Image edit LoRAs with attention bias and image strength controls, a MOSS-specific model merge node, rope scaling for MOSS-Image, native Fibo 2.0 support, and single-head LoRAs applied to MiniMax H3's PDD head banks. Convert Image Color Space and Save Image (Advanced) can now write ACEScg EXR files, the new Pad Image node pads with mirroring or to a multiple of a size, and a Branch node lands in core. Paid partner nodes add Kling 4.0, PixVerse V6-Max, FLUX 3 Max, 2K and 4K for FLUX 3 Video, Ideogram Vector SVG, Bria FIBO 2.0, Meshy T2.5, VEED Lipsync 2.0, Mirelo SFX and a Crystal Video Upscaler.
Worth knowing
Partner nodes run on the companies' paid APIs through Comfy credits, not on your GPU. The release also turns on PyTorch attention in the VAE on AMD cards, lowers MOSS-Image's peak memory, and updates the workflow templates to v0.11.97 and the frontend to 1.57.4.