Meta Muse Image Comes to ComfyUI as Official API Nodes

Muse Image, Meta's most advanced image generation model from Meta Superintelligence Labs, is now available in ComfyUI through two official API nodes: Meta Muse Image Text to Image and Meta Muse Image Edit (Meta announcement, ComfyUI PR).

Meta describes Muse Image as its most advanced image generation model yet: it follows instructions faithfully, edits with precision, and composes from multiple references, with agentic tool use capabilities integrated into Meta's Muse Spark ecosystem.

Muse Image announcement visual from the Meta AI blog

What the ComfyUI nodes offer

Both nodes run the muse-image-1.0 model through ComfyUI's partner API system:

• Text to Image: prompt-based generation with 14 selectable aspect ratios and a reasoning_strength control (high or low) that trades prompt adherence against speed.
• Image Edit: accepts up to 10 input images in a single edit pass, so multi-reference composition and large multi-element edits can run in one node. Per-tool enablement lets you switch individual edit capabilities on or off.

The official Muse Image edit template from the ComfyUI template gallery

The official Muse Image text-to-image template

How to use it

Update ComfyUI to the latest build, then find both nodes in the Node Library under the Meta partner section, or load them from the Templates panel. Generation runs on Meta's servers through the API node system: no local weights download is required, and billing follows the standard partner API pricing.

With Muse Image, ComfyUI's partner roster now covers image generation from Meta alongside the existing text model nodes, and the 10-image edit input is currently the largest multi-reference capacity of any image edit node in ComfyUI.