Tested on Anima-Base v1.0: no prompt hit the 512-token cap, cosine similarity moved the wrong way, and only the DiT-side conditioning changed who got drawn.
Tested 11 character combos × 3 seeds on Anima-Base: trigger-only prompts passed 3/33, full prompts 8/11, and copying one training-caption phrase fixed the height gap.
Measured on M1 Max 64GB: weight-only INT8 runs 41% slower than fp16, a hand-written Metal int8 GEMM 15.6x slower, MPSMatrix int8 3x, and MLX 8-bit 9% slower.
Tested on M1 Max 64GB ComfyUI: v0.24.1 fails to load int8_tensorwise, v0.30.1 hits the missing aten::_int_mm MPS kernel, and a dequantize patch runs slower than bf16.
Tested on M1 Max 64GB: AnimaLoraToolkit + Anima-Base crashes on torch 2.6.0 (MPS SDPA bug) but runs 300 steps clean on 2.7.1, at ~21 s/step, roughly 9x slower than an RTX 5090.
Muon and SOAP cut final diffusion loss 18% vs AdamW. Mapping that to Anima (Cosmos DiT) and SDXL U-Net LoRA training, plus existing LoRA-Muon research, before writing any code.
Merged a 4th girl with a makeup toggle into one Anima LoRA (518 images, rank256, 21.5h on RTX 5090). Epoch pick vs design bleed, why makeoff fails in multi-girl prompts, 6/6 one-shot.
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).
Three LLMs converted the same 10 Japanese scene briefs into Anima (Qwen-DiT) prompts, generated as 60 fixed-seed images on an M1 Max with a merged 3-character LoRA. The Qwen-to-Qwen affinity hypothesis did not survive; a strict formatter brief with character-count locks is what actually moved the results, and two failure modes survive any prompt.
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.
