| --- |
| license: other |
| base_model: krea/Krea-2-Turbo |
| base_model_relation: adapter |
| pipeline_tag: text-to-image |
| tags: |
| - lora |
| - krea-2 |
| - krea-2-turbo |
| - comfyui |
| - character |
| - ai-toolkit |
| widget: |
| - text: "Kaitlyn, portrait photo" |
| --- |
| |
| # Kaitlyn Krea 2 Turbo LoRA v1 |
|
|
| A character LoRA for **Krea 2 Turbo**, trained with [AI Toolkit](https://github.com/ostris/ai-toolkit). |
|
|
|  |
|
|
| ## Usage |
|
|
| - **Trigger word:** `Kaitlyn` |
| - **Recommended starting strength:** `1.0` |
| - **Base model:** [`krea/Krea-2-Turbo`](https://huggingface.co/krea/Krea-2-Turbo) |
| - **Format:** BF16 Safetensors LoRA, rank 32 |
|
|
| Place `kaitlyn_krea2_turbo_v1.safetensors` in your ComfyUI `models/loras` directory, load it with your Krea 2 Turbo workflow, and include `Kaitlyn` in the prompt. |
|
|
| ## Training details |
|
|
| - Trainer: AI Toolkit 0.12.26 |
| - Training steps: 3,000 |
| - Embedded base model identifier: `krea2` |
|
|
| ## Preview provenance |
|
|
| The preview image was generated in ComfyUI with the LoRA at strength `1.0` using a Krea 2 Turbo Q5 GGUF workflow. The workflow also had `krea2filterbypass3.safetensors` enabled at strength `1.0`. |
|
|
| ## Notes |
|
|
| This repository contains the LoRA adapter only. Download the Krea 2 Turbo base model separately and follow its license and usage terms. |
|
|
