Somni-ComfyUI Serves Up A Slick Frontend For ComfyUI On Any Device

    
        By vramkickedin    
     | 
    
            May 27, 2026 at 12:34 pm        
    
     | 
    
        2 min read

The somni-comfyui project by searcc delivers a polished, streamlined web interface for ComfyUI that works on desktop and mobile. It offers a Gemini-style chat mode for quick generations and a Pro mode with full control over samplers, schedulers, LoRAs, and CFG. Your existing workflows and models stay intact because it communicates with ComfyUI over HTTP without moving anything.

Searcc designed somni to run alongside any ComfyUI installation, giving users an easier frontend that still exposes all the backend’s capabilities. It launches as a tiny Python proxy—no background services—and closes when you’re done. This makes it a practical choice for anyone who wants a guided, responsive UI without rebuilding workflows.

Easy mode and pro tools in one

Key Features

Chat-style easy mode for quick generations.
Pro sidebar with sampler, scheduler, LoRAs, CFG.
IP-Adapter: General, Face, and FaceID modes.
Batch generation with scrollable preview.
Gallery swipes on mobile, arrows on desktop.
No background services—runs on demand.

Serious hobbyists and small agencies will appreciate generating images with a single prompt or diving into advanced settings for fine‑tuning. The mobile‑friendly design means you can queue generations from your phone on the same Wi‑Fi network. Privacy‑conscious professionals benefit from running everything locally without any data leaving their machine.

Roadmap and technical simplicity

Installation is via an installer.bat that points to ComfyUI and writes two launch scripts. Currently Windows-only, but Linux and macOS scripts are on the roadmap, along with multi-image IP‑Adapter, workflow presets, and inpainting. The UI is one HTML file with no build step or npm, so you can open and tweak it directly.