HyperGAN
Built from HyperGAN's public GitHub profile. Not on Take My Repo yet, so there are no posts or updates here.
Signed concept LoRA training and inference controls for OpenMOSS MOSS-Image-SFT, adapted from particle-sliders
Ming-Image Design-Layer particle sliders — product repo (Anima/Krea2 pattern); base inclusionAI/Ming-Image-0.1-Design-Layer
Stabilizing GANs by replacing known priors with learnable particle clouds.
Shared ParticleGAN training, adapter controls, checkpoints and search for model-specific concept sliders.
Human-readable Network Definition Language - a DSL for defining neural networks
Supra2-IMG concept sliders: 1,600 steps in 134.4s on RTX A6000, with 6.8 MB LoRAs and 3.4 MB rank-8 distills.
Define a Jev quality criterion, curate successful internal states, train a ParticleGAN adapter, and apply it. Tiny Aya is the worked example.
Krea-2 turbo-bbox particle sliders — product repo (Anima pattern); base jimmycarter/krea2-turbo-bbox
Anima Candlelit and Moonlit particle sliders, distilled LoRAs, ComfyUI plugin, and reproducible training
Music 3 particle sliders — product repo (Anima/Krea2 pattern); shared particle-sliders-core
Paired-context ParticleGAN 0.10 behavior learning for the Gemma 4 Developer Agent competition
Muse Glimmer particle-slider integration using upstream ParticleGAN and ridge LoRA distillation
Native Qwen3.5 particle teachers, LoRA distillation and validated Delta League weight edits
Experimental LTX-2.5 source extraction; historical UNI and dummy-only modern preview, no native/GPU/media qualification
Particle Sliders for Qwen-Image 2.1 using the shared HyperGAN core. Tiny native CPU tested; full-model quality unvalidated.
Source-only Z-Image shared-game CPU preview extracted for review from HyperGAN/particle-sliders
Experimental Flux packed prerequisites; shared training admission blocked; no pretrained or GPU qualification