ROBOMAR ONE: A Unified Local Frontend for ComfyUI
robomar_ai_art · reddit · 2026-07-15
The author is developing ROBOMAR ONE, a local desktop frontend designed to unify image, video, editing, and upscaling workflows in ComfyUI.
Rather than replacing ComfyUI, it acts as an upper-layer interface: users export their API workflows from ComfyUI, and the app automatically identifies the workflow type and editable nodes. It then generates corresponding UI controls to pass parameters, prompts, and reference images back to the local ComfyUI API for execution. Results are returned to a built-in gallery that manages both images and MP4 files.
Currently supported features include:
- Local ComfyUI connection and API workflow import
- Auto-detection of parameters for different models/workflows
- Distinct control panels for image, video, and upscale tasks
- Reference image upload and node mapping
- Prompt generation and rewriting
- A/B comparison, zoom, full-screen preview, and output file deletion
- Independent saving of workflow configurations to prevent overwrites
The author emphasizes that original workflow files are never modified; all compatibility fixes occur strictly in temporary copies.
Related event: ROBOMAR ONE: Unified Local Frontend for ComfyUI(2 posts)→
More from Multimodal
- ComfyUI Krea2 Prompt Styler Node Revived in a Fork With a Massive Style Library — Winter_unmuted · 2026-10-27
- Single HTML file generates retro poster designs in-browser with zero LLM calls — Afinetheorem · 2026-10-27
- A Redditor Used Copilot to Imagine and Archive Their Own '80s/'90s Winter Streetwear Concepts — Valuable-Sky2134 · 2026-10-27
- Transkun open-source piano transcription model nails MIDI, sustain events included — alexanderchen · 2026-10-27
- fal adds FLUX 3 Image Klein LoRA training: 10 reference images, 3 LoRAs per request — OdinLovis · 2026-10-27
- Guy turns his girlfriend's selfie into a playable AI game ad about popping pimples — CurieuxExplorer · 2026-10-27