← sources

[Hiring] ComfyUI expert to build a panel-sequence workflow for a 200-page graphic novel

source: reddit (reddit:comfyui)published 2026-09-24 17:30ingested 2026-09-25 02:42

retained text

Hey all,

I'm producing a 200-page graphic novel based on a bible story, and I'm looking to hire an experienced ComfyUI person to build the workflow I'll use to actually make it.

The one thing I care about most: **I want to generate a panel, then continue the same scene and generate the next panel — and the next — while keeping the characters, environment, and art style consistent from one panel to the next.** That panel-to-panel continuity is the whole ballgame. Everything runs in the cloud.

**What I'm looking to have built:**

* **A sequential panel workflow** (the main thing) — generate a panel, then feed it forward to create the next beat in the same scene (new angle, new pose, closer/wider shot) while characters and background stay consistent.
* **A custom style LoRA** — trained for a specific painterly, fine-art realism look (master-painter feel, oil-on-canvas texture, cinematic lighting). Something custom or merged — not just a stock LoRA pulled off Civitai.
* **ControlNet / OpenPose** — so I can direct poses and composition and hold staging consistent as a scene moves panel to panel.
* **An edit-model + inpainting workflow** — to refine and repair: fix drift, adjust or swap backgrounds, correct details, and keep characters/style consistent when a panel isn't perfect first try.

I'm imagining this as **3–4 clean, organized ComfyUI workflows** you hand off to me so I can run the whole book myself.

**Deliverables:** the working cloud workflows, a few test panels showing a short sequence from one of my actual scenes (to prove the continuity), and a short handoff call + notes so I can operate it all solo.

I'm open to suggestions on models and methods — I've been getting strong results from GPT image generation and want to talk about where a commercial model fits alongside the ComfyUI pipeline. I've also got detailed character reference sheets and a locked art style ready to share.

**Important:** this is a genuinely hard project, so please only reach out if you've done this kind of thing before and are confident you can pull it off. I'm looking for a real expert in AI image generation, not someone learning on my dime.

If that's you, comment or DM me with examples of relevant work — especially anything showing character/style consistency across multiple images — and your rough rate. Happy to share more about the project from there. Thanks!

document fetched 2026-09-25 00:45 · canonical https://reddit.com/r/comfyui/comments/1wp8005/hiring_comfyui_expert_to_build_a_panelsequence