New Solution - AnimaEdit: Reference & Instruction Image Editing LoRA
retained text
https://preview.redd.it/8nsixv6txj0i1.png?width=2116&format=png&auto=webp&s=03c59b9511f27f2d108f5ebba9a7f0cb9de75c1e True image edit for Anima. Based on \`ElectronGirl\` solution. Link: [https://civitai.com/models/3041608/animaedit-reference-and-instruction-image-editing-lora](https://civitai.com/models/3041608/animaedit-reference-and-instruction-image-editing-lora) https://preview.redd.it/sdfythdnyj0i1.png?width=1292&format=png&auto=webp&s=bb3806ad4cd97886a55fb4530cef8a40354c1b9f And, there is v2 update for \`Anima-Role-Ref-KV\` solution with better character resemblance. Link: [https://civitai.com/models/3021867/anima-role-ref-kv](https://civitai.com/models/3021867/anima-role-ref-kv) https://preview.redd.it/nfhc7z3nzj0i1.png?width=1664&format=png&auto=webp&s=86a1bb77cf6f8395356b20acee4ceb38acf48d5e And, let's not forget about new version of SnapRef v2a! Link: [https://civitai.com/models/3028191/snapref](https://civitai.com/models/3028191/snapref) Well... Happy to see Anima continues to get better and better! But 0.6B textencoder is holding it back!
document fetched 2026-11-10 00:30 · canonical https://reddit.com/r/StableDiffusion/comments/1xrmz20/new_solution_animaedit_reference_instruction