If you like this model, remember to give a like to / follow <a target="_blank" rel="ugc" href="https://civarchive.com/models/135477/molkeun-mix?modelVersionId=3392495">the original</a>!<a target="_blank" rel="ugc" href="https://civarchive.com/user/Mol_Keun">Mol_Keun</a> released an Anime 3.8B version of their <a target="_blank" rel="ugc" href="https://civarchive.com/models/135477/molkeun-mix?modelVersionId=3392495">Mol_Keun Mix</a>. But it seems it isn't working for some people who don't use ComfyUI? So here's a conversion of it I tested and confirmed works on Forge Neo. It's also quantized to INT8_convrot to run on my potato and run faster. Update: Added the original bf16-precision denoiser without the embedded TE and VAE, people with more VRAM than I can test if it works on Neo for them.Update: Also added the non-3.8B version converted to int8_convrot since it can go on the same model page. If using a Turbo LoRA with it, you'll need to turn the strength down, 0.7 worked for me.Update: Added int8_convrot conversion of <a target="_blank" rel="ugc" href="https://civarchive.com/models/2880965/mol-keun-comix-anima">Mol Keun COMIX</a>, OGMustard's merge of Mol Keun Anima and JetlistiC's <a target="_blank" rel="ugc" href="https://civarchive.com/models/1999215/maturacomix">MaturaComix</a>. Get Anima's Qwen3 0.6B text encoder from their <a target="_blank" rel="ugc" href="https://huggingface.co/circlestone-labs/Anima">HuggingFace page</a>. Optionally, get the VAE there too, or use any Anima VAE. I believe Mol_Keun originally bundled <a target="_blank" rel="ugc" href="https://huggingface.co/Anzhc/Qwen2D-VAE">Qwen2D</a>, which also has an <a target="_blank" rel="ugc" href="https://huggingface.co/Anzhc/Qwen2D-Anime-VAE">Anime</a> version. All credit to the original merge author Mol_Keun. <s>If they don't want quantizations like this posted just dm me and I'll take it down.</s> Wow they featured it on their reccomended resources, guess they approve! Glad I could help get their checkpoint into the hands of Forge Neo users! =M Apologies for the low quality showcase images, I'm on a GTX 1060. Mol_Keun's recommended settings: Sampler: Euler a<ul><li>CFG Scale: 4~6</li><li>Steps: 25~40</li><li>Clip Skip : 2</li></ul>Tested settings for Turbo LoRAs: (with <a target="_blank" rel="ugc" href="https://civarchive.com/models/2895815/anima-29b-turbo-v29-remap">Anima-2.9B-Turbo-v2.9-remap</a>, probably works with <a target="_blank" rel="ugc" href="https://civarchive.com/models/2857451/anima-29b-turbo-lora-remapped">Anima 2.9B Turbo Lora Remapped</a> too)Samplers: ER SDE, Res Multistep, Euler aSchedulers: Beta, Simple, Linear Quadratic, Bong Tangent<ul><li>CFG Scale: 1-2</li><li>Steps: 8-16</li><li>Clip Skip: Whatever the default is</li></ul>

Mol_Keun Mix Anima 3.8B int8_convrot

int8_convrot conversion of <a target="_blank" rel="ugc" href="https://civitai.red/models/2880965/mol-keun-comix-anima">Mol Keun COMIX</a>, OGMustard's merge of Mol Keun Anima and JetlistiC's <a target="_blank" rel="ugc" href="https://civitai.red/models/1999215/maturacomix">MaturaComix</a>.

Files with SHA256: 3360716655d6a7911ee6819f659ac21971fd97a04bcea5131f0dc621df9b7621