The Tools I Used to Create My Graphic Novels

Creating a graphic novel no longer means starting from a blank page. For this project, I combined AI-generated imagery with a set of free, open-source tools: GIMP, Inkscape, and Scribus.

AI helped generate the raw visual material—but these tools are what turned that material into a coherent, readable, print-ready book.

Think of it less like drawing every panel by hand, and more like directing, editing, and assembling a film.

GIMP: Shaping AI-Generated Images into Panels

Most of the visual work happened in GIMP—not by painting from scratch, but by refining AI-generated images into usable comic panels.

Here’s what that looked like in practice:

  • Resizing and scaling images to match consistent panel formats

  • Cropping and reframing to improve composition and focus

  • Layer-based layout work to organize panels and visual flow

  • Colour correction (levels, curves, contrast) to unify tone across scenes

AI images are unpredictable. Lighting shifts. Perspectives vary. Characters drift off-model. GIMP became the place where those inconsistencies were corrected.

Colour grading was especially important. Subtle adjustments helped maintain a consistent atmosphere across pages—so the story felt continuous rather than stitched together.

This stage felt closer to editing than illustrating. The images existed—but they needed to be shaped into something intentional.

Inkscape: Imposing Structure on AI Output

If GIMP was about shaping images, Inkscape was about imposing order.

AI doesn’t think in panels. It doesn’t understand pacing or page structure. That had to be built separately.

Inkscape handled:

  • Panel borders and page grids

  • Speech bubble placement and guides

  • Alignment and spacing across pages

  • Clean overlays that sit on top of raster artwork

Because it’s vector-based, everything remains sharp and adjustable. That’s critical when layering structured elements over AI-generated images, which can vary in resolution.

This is where the comic language really emerges—the rhythm of panels, the spacing, the readability. Without this step, the images would remain just that: images, not a story.

Scribus: Turning AI Artwork into a Finished Book

The final stage took place in Scribus, where everything was assembled into a complete, print-ready graphic novel.

Here, the focus shifted again:

  • Sequencing pages to control pacing and narrative flow

  • Placing text and refining typography

  • Setting margins, bleeds, and trim for print

  • Exporting a press-ready PDF

At this point, the AI-generated origins of the artwork matter less. What matters is whether the book reads well—whether the eye moves naturally, whether the text sits comfortably, whether the pages feel balanced.

Scribus is where all of those decisions come together.

Why This Workflow Works

AI can generate images—but it doesn’t create a finished comic.

That requires:

  • Selection

  • Editing

  • Structure

  • Restraint

This workflow separates those responsibilities clearly:

  • GIMP shapes and unifies raw AI imagery

  • Inkscape imposes structure and readability

  • Scribus delivers a professional final product

Each tool compensates for something AI doesn’t do well.

Final Thoughts

There’s a misconception that AI replaces the creative process. In reality, it shifts it.

The work isn’t eliminated—it moves.

Instead of drawing every line, you:

  • Choose what to keep

  • Decide what to change

  • Control how everything fits together

The result is still a crafted object. Still a series of deliberate decisions.

The tools made it possible.

AI made it faster.

But the final book still had to be built—panel by panel, page by page.

Next
Next

Hieronymus Bosch and the Illustrious Brotherhood of Our Blessed Lady