🍃 Nick & Charlie | LoRA — Heartstopper | Base: ANIMA

Captures the iconic hand-drawn art style of Alice Oseman's Heartstopper webcomic and brings Nick Nelson and Charlie Spring to life in ANIMA.

🙏 Before anything else

This LoRA took a lot of work — picking the images, building the dataset and tags, training, and testing the checkpoints. If you have a moment, drop a Like and especially post your results in the comments (with the prompt/settings you used) — it helps me a lot to know what to improve in the next versions. 🔥

⚠️ The model was trained without any images and without any photos of the live-action actors. It is strictly based on the original graphic novel / webcomic art style. Use it respectfully.

💬 Check out the image descriptions and prompt examples on the posted preview images — they show how the prompts are structured!

💬 Leave your generations in the comments, and tag the LoRA so I can see how it's looking — it would help a lot.

📦 Trained with 92 carefully selected images in total from the original webcomic art.

📊 About this model

The main goal of this LoRA is to capture the iconic hand-drawn art style of Alice Oseman's Heartstopper webcomic and bring Nick Nelson and Charlie Spring to life in ANIMA.

While the LoRA is not 100% perfect, it works extremely well!

💬 Just to remind you, the comics have simple backgrounds and pastel tones, emphasize this in the prompt, otherwise the characters won't 'match' the rest of the work

🌀 Nick Nelson

Nick generates consistently with high quality across most poses and outfits. His facial features, hair shape, and athletic frame lock in very reliably.

🌀 Charlie Spring

Charlie's output quality depends significantly on the generation resolution. Since the original comic source drawings vary widely in scale and detail, generating Charlie effectively may require a few resolution tweaks and test runs until you find the sweet spot for your specific scene.

🎨 Using the Art Style with Other Characters

If you want to extract the Heartstopper art style without generating Nick or Charlie:

• Use the trigger tags heartstopper + webcomic in your prompt.

• Combine this with other character LoRAs to apply Heartstopper's unique hand-drawn pastel aesthetic to any character!

ℹ️ Generation tips

• Resolution Flexibility: There is no single "standard resolution" for this LoRA. Because the source art varies greatly, trying different resolutions (portrait, square, or wide 2:3 / 3:4) depending on scene context makes a huge difference in image quality.

• Fixed Seed Testing: When experimenting with resolutions or prompt tweaks, try using a fixed seed to observe how resolution directly impacts detail and character rendering.

• Recommended starting point: On ANIMA base with 12 steps + acceleration LoRA, the output keeps a clean hand-drawn webcomic feel that fits Heartstopper perfectly.

• Check the descriptions of the posted preview images — they demonstrate how to structure tags and natural language descriptions (or multi-panel layouts) for best results.

⚙️ Generation settings used for preview images

All preview images were generated using:

• Base model: Anima 1.0

• Acceleration LoRA: strength 1.0

• CLIP: anima default

• Steps: 12

• CFG: 1.0

• Sampler: er_sde - beta

• Image size: there’s no fixed resolution for this LoRA; it works at resolutions above 512x and in any format, although the performance is pretty hit or miss. I recommend testing various resolutions for your specific prompt.

👍 One more thing

I'm just getting started with this — if you have a moment, drop a Like and share your generations with the settings you used. It really helps me improve the next versions. Have fun! 🍃