Tested over 3 bakes on RunPod RTX 5090: a coined subtractive makeoff tag never fires at cfg 1.0, while additive earrings+makeup tags switch both ways with the face unchanged (ep140).
Tested arXiv 2607.01232's layer localization under SFT: Qwen3-4B depth 25/50/75% vs all-layer LoRA, trained on a RunPod 4090. The eval-loss U-shape is real; the rewrites disagree.
Fine-tuned Qwen3-4B on 799 of my own edit pairs, quantized to a 2.3GB GGUF at 38 tok/s on an M4 mini. Eval loss looked fine, but it barely removed slop — and the real fix was feeding paragraphs, not single sentences.
Merged kei, kana, and koharu into a single Anima (Qwen-DiT) LoRA and ran my first training on Blackwell (RTX 5090, sm_120). Hands-on log: the cu128 / torch2.8 / SDPA stack swap from the 4090, why the weakest character gets absorbed (caption asymmetry, not rank), and how trigger-only prompts separate three close-packed characters at ep143 without ControlNet.
Tested on M1 Max ComfyUI: newly free JANIMA vs Hexer Minimal Toon Anima V1 vs anima-base, one character LoRA, same seed. Hexer keeps the outfit; JANIMA adds clothes but draws the quietest backgrounds.
rank128 + 20 two-character images killed the v1 ahoge bleed and body fusion on this Anima dual-character LoRA. Lap-sit stays a Qwen3 text-encoder limit; sweet spot is ep140.
One Anima (Qwen-Image DiT) LoRA, two characters, trained on RunPod: can they touch? Hugs and piggyback hold, lap-sit fuses; stacked limbs survive, interleaved break. Best at ep100, Turbo.
Hexer Minimal Toon's new Anima V1 is a DiT checkpoint, not an SDXL one: 4.1GB bf16, separate VAE/text-encoder ComfyUI folders, Anima-only LoRA, and a non-commercial license the Civitai page doesn't show. What changes vs Illustrious v3.1 before you load it.
On Anima-Base, my character LoRA bent its legs even on standing. Adding upright references didn't fix it; cutting 36 posed full-body images did. Subtract, don't add.
