Krea2 Character LoRA Collection
A collection of 84 character LoRAs for the KREA2 base AND turbo model, each paired with a persona .md file describing the character's stable facial features, variable appearance, and expressions.
Every LoRA ships as <Name>_KREA2.safetensors alongside <Name>_KREA2.md inside its own folder under Characters/.
Updates
Repository restructure
- All characters moved under a single
Characters/directory (one folder per character containing the.safetensors, the persona.md, and the training captions). - A new
Pruned_Comparisonfolder was added with one preview image per character (see below).
Dynamic re-rank + block pruning
Every LoRA was re-ranked with dynamic sv_fro SVD (retention 0.998, max rank 16) and block-pruned by dropping the 8 lowest-impact transformer blocks (blocks 1-8, ~3% of delta energy).
Retrained character
- Jung Ho-yeon (
Ho-yeon) was retrained on a new dataset, then re-ranked and pruned like the others.
Fully automated pipeline
This repository - dataset, training, and upload - was produced end-to-end by an autonomous Hermes Agent (served by the Kimi K3 model) from a natural-language request. No manual review gates anywhere in the loop:
- Scraping - automated image collection per character from web photo pools.
- Cropping & curation - face/head detection Ultralytics Yolo and crop, with a hard content filter (No AI generated images, correct subject only, sharpness threshold).
- Aesthetic gating - every candidate image scored by a discriminator built on the ERNIE aesthetic image scorer; low scorers rejected automatically with a hard pass/fail threshold.
- Captioning - datasets captioned with Gemma-4 26B (abliterated) vision MTP for unrestricted, descriptive captions, fast (1.2s/i).
- Persona synthesis - each
<Name>_KREA2.mdgenerated from caption trait-analysis (facial features, range of expressions). - Infrastructure - training pods created, configured, monitored, and torn down automatically via the RunPod API (MCP).
- Training - LoRAs trained on pods with Ostris AI-Toolkit with user predefined settings.
- Publishing - weights and persona files uploaded to this repo by the agent.
Pruned comparison
One example image per character, generated with the pruned LoRA. Click a character to expand.
Adria Arjona
Adriana Lima
Alexandria Ocasio-Cortez
Alix Lapri
Amelia Gray
Ana De Armas
Anne Hathaway
Anya Taylor-Joy
Ariana Grande
Autumn Falls
Barbara Palvin
Bella Poarch
Billie Eilish
Camila Cabello
Camille Cerf
Camille Hurel
Cara Delevingne
Carla Poquin
Cindy Kimberly
Corinna Kopf
Daniela Avanzini
Demitra Kalogeras
Doja Cat
Dorozea
Dua Lipa
Eiza González
Elydia
Emilia Clarke
Emily Ratajkowski
Eva Mendes
Fernanda Ly
Gal Gadot
Hailey Rose
Hannah Kleit
Ho-yeon
Ice Spice
Imaan Hammam
Inde Navarrette
Irina Shayk
Jessica Alba
Kaia Gerber
Kelsey Merritt
Kendall Jenner
Kylie Jenner
Lara Rajagopalan
Lauren Mayberry
Lea Elui
Leah Halton
Lena Gercke
Lena Mahfouf
Léa Meniel
Madalina Ghenea
Madison Beer
Malu Trevejo
Manon Bannerman
Margot Robbie
Megan Fox
Megan Skiendiel
Monica Bellucci
Natalie Portman
Nicki Minaj
Nikita Bellucci
Olivia Cooke
Pokimane
Rihanna
Rima Hassan
Rosalia
Sab Zada
Scarlett Johansson
Shakira
Shay
Sofia Jamora
Sofia Richie Grainge
Sommer Ray
Sophia Laforteza
Sydney Sweeney
Tinashe
Tokyo
Tyla
Valkyrae
Wejdene
Yoonchae Jeong
Zendaya
Zoe Kravitz
Known limitations
- Trained for likeness; style flexibility depends on your prompt and sampler settings.
Disclaimer
These LoRAs reproduce the likeness of real people and are shared for research and creative purposes. Please use them responsibly and respect the rights of the individuals depicted - don't create misleading, harmful, or non-consensual content. By downloading, you take responsibility for what you generate.
Credits
- Pipeline automation: Hermes Agent (Nous Research) on Kimi K3
- Training: Ostris AI-Toolkit
- Aesthetic scoring: ERNIE aesthetic scorer (CRT-Nodes ComfyUI packaging)
- Captioning: Gemma-4 26B (abliterated) vision MTP
- Infrastructure: RunPod
- User Discord https://discord.gg/uV6ZNgg33j