tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- output:
url: images/0_nOhuo3uGFo0-Cucr.jpg
text: '-'
- output:
url: >-
images/08_spider-diagram-brainstorm-whiteboard-in-pastel-blue-pastel-green-pink-spaced-color-blocks-style.webp
text: '-'
- output:
url: images/class-diagram-dependency-relationship.jpg
text: '-'
- output:
url: images/istockphoto-2114257458-612x612.jpg
text: Circle chart infographic template with 3 options for presentations
- output:
url: images/Mindmap-created-using-Cacoo-1024x712.png
text: '-'
- output:
url: images/venn-diagram.jpg
text: '-'
- output:
url: images/what-is-venn-1.jpg
text: '-'
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: cognitrain, cognitrain logic, cognitrain structured
license: creativeml-openrail-m
CogniTrain – Instruction-Aware Training & Reasoning LoRA
- Prompt
- -
- Prompt
- -
- Prompt
- -
- Prompt
- Circle chart infographic template with 3 options for presentations
- Prompt
- -
- Prompt
- -
- Prompt
- -
Model description
CogniTrain is a LoRA model designed to improve instruction awareness, semantic alignment, and structured concept representation during generation.
Instead of focusing purely on visual style, CogniTrain emphasizes: • Clear interpretation of complex prompts • Strong concept-to-output alignment • Reduced randomness in structured generations • Better control over multi-object and multi-context scenes
The model is optimized for creators who value precision, logic, and prompt reliability over purely artistic randomness.
Recommended strength: 0.5 – 0.85
Trigger words
You should use cognitrain to trigger the image generation.
You should use cognitrain logic to trigger the image generation.
You should use cognitrain structured to trigger the image generation.
Download model
Download them in the Files & versions tab.
• Technical illustrations • Product concept visualization • Educational and explainer visuals • UX / system diagrams • Multi-object structured scenes