--- license: cc0-1.0 base_model: - black-forest-labs/FLUX.1-dev pipeline_tag: text-to-image tags: - art - glitch - 3D - character - image --- Resized image # **GlitchHunter** **GlitchHunter** is Flux dev Lora trained on my outputs from MJ and SDXL. Focus is on unusual erratic characters. It is trained on few types of styles and merges them really well into new forms with Flux and carries style into other subjects. Biased on neon cyberpunk comic style. Training data includes part of my [Supernatural Creatures animation collection](https://magiceden.io/launchpad/supernatural_creatures). You can find more info about the collection and the process behind it in [this medium article](https://medium.com/@glitchcandies/supernatural-creatures-fcb993254d04). GlitchHunter Lora is trained on 52 images. Dataset styles and output include: - designer toy inspired mecha sculptures for 3D image style - toon shader / sprite like 2.5D style - abstract comic hand-drawn 2D style To get those styles type in "3D", "comic", into prompts. It was captioned using gpt-4o-mini and my manual input. It likes simple prompts and transfers styles pretty well. Prompt example: *A vivid, surreal mecha creature with colorful flora and mushrooms with whimsical shapes emerges, blending nature, metallic like monster, 3D imagination in harmonious chaos.* **Settings**: CFG: 1, Steps: 28 or 38 work well Trigger word: **[GlitchHunter]** Trained using [Glif Loradex](https://huggingface.co/glif/loradex-flux-dev) while beta testing the platform. Licensed CC0. You can try it on [Glif in GlitchCandies Creature Maker](https://glif.app/@Koshi/glifs/cm16e0vir0005dlu6lp9oufai) and GlitchCandies Creature Maker with image and text. ## Sample Outputs
3D Style 2.5D Toon shader like Hand-drawn Style
3D Sample 1 2.5D Sample 1 Hand-drawn Sample 1
3D Sample 2 2.5D Sample 2 Hand-drawn Sample 2
--- license: cc0-1.0 ---