krea2 furry
The model was trained on approximately 1600 images of various types. To ensure versatility, it was trained on multiple styles, and secondary images were assigned lower probabilities.
Model source
- Kind
- LORA
- File size
- 327 MB
- Base model
- Krea 2
- Version
- vkrea2_furry_0716
- Updated
- Oct 26, 2026
- Downloads
- 1.8K
- Source
- Civitai
Source description
Perform longer training on a larger-scale dataset.
However, the different concepts are interwoven with each other and the learning difficulty varies. The model training might be overly intense and a lower intensity might be needed.
===
在更大规模的数据集上进行更久的训练
但是不同概念之间相互交错和学习难度不同,模型训练有些过度,可能需要低一点的强度.
Sources
1 source
Sources
1 sourceModel artifacts
1 artifact
Model artifacts
1 artifactkrea2_furry_0716.safetensors
safetensors · BF16 (unverified) · 327 MB · SHA-256 adcd03cd80ee…de15 · Civitai
Download fileSource excerpts
2 excerpts
Source excerpts
2 excerptsSource context: 2485 downloads · Type LORA · Base model Krea 2
Source context: 1252 downloads · Type LORA · Base model Krea 2
