ArtificialGuyBR

Home / Blog / Training LoRAs

Doodle Style Redmond LoRA: Hand-Drawn Sketch Art for Qwen-Image

Generate consistent doodle style illustrations with this Qwen-Image LoRA. Trigger words, usage tips, and examples.

2 sources cited Training LoRAs

On this page

  1. What This Model Does
  2. Trigger Words
  3. How to Use
  4. License
  5. Common Errors and Fixes
  6. FAQ
  7. Sources
  8. Related Models

Doodle Style Redmond LoRA for Qwen-Image

TL;DR

What This Model Does

Doodle Style Redmond is a Low-Rank Adaptation (LoRA) trained on Qwen/Qwen-Image-2512. According to the model card, it generates "high-quality doodle style content with excellent consistency." The Civitai description adds that the LoRA has a high capacity to reproduce doodle style in a wide variety of themes, making it a versatile style adapter rather than a single-subject LoRA.

The training was made possible by GPU compute sponsored by Redmond.AI. The author, ArtificialGuyBR, distributes the model for free.

Sample output from the Doodle Style Redmond LoRA for Qwen-Image

Trigger Words

The style requires one or both trigger words in your prompt:

Example prompt: Doodle, DoodleRedm, a small robot watering a plant

How to Use

The author recommends ComfyUI for best results:

  1. Load Qwen/Qwen-Image-2512 as the base checkpoint
  2. Add the LoRA with a Load LoRA node
  3. Include the trigger words in your positive prompt
  4. Generate

The LoRA loads the same way through the Hugging Face diffusers integration:

from diffusers import QwenImagePipeline

pipe = QwenImagePipeline.from_pretrained("Qwen/Qwen-Image-2512")
pipe.load_lora_weights("artificialguybr/Doodle-Redmond-QWENIMAGE")

image = pipe("Doodle, DoodleRedm, a tiny robot watering a plant").images[0]

The LoRA is distributed as a Safetensors file. The model card does not document special sampling parameters, so standard Qwen-Image settings (native 1024×1024 output) apply.

Sample output from the Doodle Style Redmond LoRA for Qwen-Image

License

Apache 2.0 — free for commercial and non-commercial use, with attribution.

Common Errors and Fixes

Error | Cause | Fix | Source

|---|---|---|---

LoRA not applying | Wrong base model | Use Qwen/Qwen-Image-2512 exactly | HF model card tags Style not appearing | Missing trigger words | Add "Doodle" and/or "DoodleRedm" | HF model card

FAQ

Q: Does this work with SDXL or Flux? A: No. This LoRA is trained for the Qwen-Image-2512 architecture.

Q: Who created this model? A: ArtificialGuyBR, with GPU sponsorship from Redmond.AI.

Q: Is it free to use commercially? A: Yes — the license is Apache 2.0.

Sources