海外のローカル生成AI情報などを日本語でまとめたり、制作物を投稿したりします |画像:MOSS-Image,Anima,Krea2|動画:MiniMax H3|ローカルLLM|気になるものは自宅のRTX 5090 / RAM64GBで試します
Pika Labs unexpectedly dropped Pika-H3-Flex—a 6-step distillation o...
SnapRef: one-image reference adapter for Anima (asymmetric attention + reference K/V cache). Training, data generation and evaluation code. - ArcAnima/SnapRef
Explore this post and more from the comfyui community
Slopus — 0.4.0 Changes since 0.3.2. What's new Image generators: MOSS Image, Fibo 2.0 Base/Edit Turbo, and Krea 2 Turbo/Raw, with model-specific guidance, negative prompts and editing controls...
Visual image editing, Anima regional prompting, and MOSS-Image layout prompting for classic ComfyUI. - EBEBEEBE/comfyui-lucid-vrp
Image-to-Video model, built with MiniMax H3. Download, explore files and run it on Hugging Face.
Text-to-Image model with 5B parameters. Download, explore files and run it on Hugging Face.
trigger word for Anima, Krea2 and MiniMax H3 : aziib pixel style trigger word for Z-image Turbo: aziib_pixel_style trigger word for flux: in the st...