Skip to content

← The Lab

ComfyUI lands in ChatGPT, and v0.44 adds ProRes saving and MiniMax H3 edit nodes

On Nov 4, Comfy announced that Comfy Cloud is now an official ChatGPT plugin, so you can run ComfyUI image, video, audio and 3D generation from a chat without installing anything or owning a GPU. The v0.44.0 core release that shipped the night before adds ProRes output and a batch of video nodes.

The plugin. Per Comfy, ChatGPT can now call Comfy Cloud directly: describe what you want, and it runs the workflow in the cloud and returns the result. It uses the same Comfy Cloud MCP connection Comfy already offered for other agents, so the ChatGPT plugin is a new way in rather than a new engine. Usage runs on your Comfy Cloud account, so check credits before you hand it long video jobs.

What's in v0.44.0, per the release notes. Video saving now supports ProRes and different chroma subsampling, which matters if your ComfyUI renders go straight into Premiere, Resolve or After Effects. MiniMax H3 gets Max partner nodes for Recast, Extend, Reframe, Insert, Lip Sync and Camera Control, plus native support for its latent upscaler and context windows. Also new: a Pruna P-Video Upscale node, 1080p/2K/4K targets on BFL's Flux Video Upscale V2, a VAE compute-dtype node and a Topaz Wonder 4 image option.

Our take: the ChatGPT plugin is mostly an on-ramp for people who never opened a node graph. Studios already using ComfyUI will care more about ProRes saving, which cuts out an ffmpeg transcode, and the H3 edit nodes for shot fixes.

Related tools on ReelStack

Skills & repos in our directory

More from the Lab

Topics: ComfyUI · Open-weight video & audio models · Video MCP servers