General descriptionStyle LoRA based on illustrations by iwamushi (岩虫).Rough uneven linework, deep shadows, real paper grain, muted print colors.Available for four bases — pick the version that matches your model:<ul><li>Anima v10 — trigger @iwamushi (new, recommended)</li><li>Anima v8 — trigger @iwamushi (previous version, flatter and warmer)</li><li>Anima v5 — trigger @iwamushi (previous version, darker and rougher)</li><li>Illustrious — trigger iwamushi</li><li>Pony — trigger iwamushi</li><li>Krea 2 — trigger @iwamushi</li></ul>This is a style, not a character — name the character you actually want.Not an official model by the artist.──────────────────────────────────────────────────────About this version — Anima v10Trigger: @iwamushi at the very start, then regular Danbooru tags. No style words needed.Prompt template:@iwamushi safe, masterpiece, best quality, score_7, &lt;your tags&gt;Negative:worst quality, low quality, lowres, artist name, signature, text, (bad anatomy:1.2), (bad hands:1.1), missing finger, (extra digits:1.4), fewer digits, disfigured, mutation, 4 fingers, (6 fingers:1.4)LoRA strength: 1.0Recommended:<ul><li>DPM++ 2M SDE, simple, 25 steps, CFG 5, 832×1248 (2:3)</li><li>Hires fix 1.5×, lanczos, 20 steps, denoise 0.35 — use it. The paper grain and the rough line come out at the higher resolution; without hires the picture looks noticeably simpler.</li></ul>Color controlThe default palette is cool and nocturnal — great for rain, snow, neon and night scenes. You can steer it:<ul><li>Warm and brown scenes: describe a warm light source (morning light, sunset, lamp light) and add (warm colors:1.4), brown theme.</li><li>Dark red scenes: add warm colors, red theme.</li><li>Night, rain and winter scenes need nothing — leave the colors alone.</li><li>Lowering the LoRA strength barely changes the color, so keep it at 1.0.</li></ul>The prompts of every showcase image are attached, so you can see where the color tags are used and where they are not.Tips<ul><li>rough uneven linework, halftone texture at the end of the prompt adds more dust and grain. Optional — some people like it, some don't.</li><li>The artist's signature can still show up now and then, most often in Re:Zero scenes (that's where the artist signs most). artist name, signature, text in the negative removes most of it — reroll the rest.</li></ul>What changed since v8<ul><li>Drawn, not flat: real light and shadow, dark masses and depth instead of flat fills and even gradients.</li><li>The artist's paper grain: uneven, chaotic texture on skin, cloth and sky — with hires it is close to the original scans.</li><li>More life in the scenes: emotions, poses and small props; scenes are composed instead of falling back to one template.</li><li>Prompts are followed as well as by base Anima, and color tags work without breaking the style.</li><li>Fewer signatures: in our side-by-side tests (same seeds and negative) v8 wrote the artist's name on several images, v10 on none.</li></ul>The last images of the gallery are side-by-side comparisons: the same prompt, seed and settings with v8 (left) and v10 (right).A weighted average of epochs 8 and 10 (80/20) of the same run — the averaged epochs keep the deep shadows of epoch 8 and add the finer detail of epoch 10.TrainingBase: Anima Base v1.0. DiT with AdaLN trained, text encoder and LLM adapter frozen.Rank/alpha 64/64 · 112 images · 1024 res, buckets 512–1536 · AdamW8bit · LR 1e-4, constant, 5% warmup · 12 epochs, release = 0.8 × epoch 8 + 0.2 × epoch 10 (exact merge of the LoRA deltas, refactored to rank 64)Trained on Anima Base v1.0 only. Behavior on Anima Turbo and third-party merges is unknown.──────────────────────────────────────────────────────Anima v8Trigger: @iwamushi (put it at the start of the prompt)Recommended style tags:rough uneven linework, halftone textureLoRA strength: 1.0Tips<ul><li>Don't add color tags like "olive muted palette" — they tint clothes olive-yellow. The LoRA already carries the palette.</li><li>Avoid the word "dirty" in texture tags — it paints stains on clothing.</li><li>If you ever see an "@iwamushi" signature in a corner, add signature to the negative prompt.</li></ul>What changed since v5<ul><li>Dataset grew from 79 to 112 images: new works added, low-res previews replaced with high-res originals, off-style pieces (neon, UI-heavy, 3D-like) removed.</li><li>Much less Re:Zero bias — characters stay on-model more often (hair color, outfits).</li><li>Noticeably more detail: backgrounds, clothing, lighting and shadows.</li><li>Handles complex scenes better: multiple characters, crowds, detailed interiors.</li><li>Warmer, sandier tone, closer to the artist's originals.</li></ul>Epoch 7 of 8 (784 steps), not the final epoch.TrainingBase: Anima Base v1.0. DiT with AdaLN trained, text encoder and LLM adapter frozen.Rank/alpha 16/16 · 112 images · 1024 res, buckets 512–1536 · AdamW8bit · LR 5e-5, constant, 5% warmupTrained on Anima Base v1.0 only. Behavior on Anima Turbo and third-party merges is unknown.──────────────────────────────────────────────────────Anima v5Previous version. Superseded by v8, which has more detail and handles complex scenes better. v5 is kept for its darker, rougher look.Trigger: @iwamushi (put it at the start of the prompt)Optional style tags: flat print color, olive muted palette, rough uneven linework, dirty halftone textureEpoch 6 of the v5 run (474 steps), not the final epoch.TrainingBase: Anima Base v1.0. DiT with AdaLN trained, text encoder and LLM adapter frozen.Rank/alpha 16/16 · 79 images · 1024 res, buckets 512–1536 · AdamW8bit · LR 5e-5, constant, 5% warmupTrained on Anima Base v1.0 only. Behavior on Anima Turbo and third-party merges is unknown.──────────────────────────────────────────────────────IllustriousTrigger: iwamushi (put it first, then regular Danbooru tags). No style tags needed.Recommended:<ul><li>LoRA 1.0 (up to 1.2)</li><li>euler_a, 30 steps, CFG 5–6</li></ul>Faces and expressionsExpression tags work, but the style draws dim eyes and shadows on the face, partly because the artist draws eyes that way, so smiles can look tired.<ul><li>Add empty eyes, shaded face to the negative prompt for brighter eyes and lighter faces. It helps most characters, not all, and can slightly change the image.</li><li>Lowering the LoRA strength doesn't help — keep it at 1.0.</li></ul>Epoch 16 of 24 (640 steps), not the final epoch.TrainingBase: Illustrious XL v2.0. UNet and text encoder trained.Rank/alpha 32/16, conv 16/8 · 79 images · 1024 res, buckets 512–2048 · batch 2 · AdamW8bit · UNet LR 1e-4, TE LR 1e-5, constant, 5% warmup · min SNR 5 · noise offset 0.03Tested on WAI-Illustrious v16, Hyphoria v0.02 and Nova Anime XL il v1.9.──────────────────────────────────────────────────────Krea 2Trigger: @iwamushi. at the very start, then describe the scene in plain sentences. No tags, no style words needed.Prompt template:@iwamushi. &lt;who or what&gt;, &lt;what they are doing&gt;, &lt;where&gt;, &lt;light and mood&gt;.Example: @iwamushi. Guts from Berserk in battered black armor sits by a small campfire at night, his huge sword leaning against a rock, embers rising into the dark, his face lit only by the fire.LoRA strength: 1.0 (0.8–1.2)<ul><li>Below 0.8 Krea starts writing "@iwamushi" in a corner — it reads the trigger as a social-media handle. The LoRA at full strength suppresses it.</li><li>Above 1.2 outfits and small details drift toward the training images.</li></ul>Recommended:<ul><li>Krea 2 Turbo, 8 steps, CFG 1, euler / simple</li><li>Native resolution, about 1 MP: 832×1248, 1248×832 or 1024×1024. No hires pass needed.</li></ul>Tips<ul><li>Atmosphere is where it shines: dusk, rain, firelight, fog, night streets.</li><li>Turbo gives very similar images across seeds — change the wording of the prompt for variety instead of rerolling.</li></ul>Epoch 6 of 12 (1344 steps), not the final epoch — later epochs start copying signatures and outfits from the dataset.TrainingBase: Krea 2 Raw. LoRA on the DiT.Rank/alpha 32/32 · 112 images, each with a long and a short natural-language caption · 1024 res, buckets · batch 1 · AdamW8bit · LR 1e-4, constant · flow shift 2.5Tested on Krea 2 Turbo (fp8).────────────────────────────────────────────────────── PonyLoRA strength: 1.0 — a balanced default for both base Pony and merges.<ul><li>Pony Diffusion V6 XL: ~0.8 if colors get heavy or red.</li><li>AutismMix and other Pony merges: up to 1.5 for a stronger style.</li></ul>Trigger: iwamushi (right after the score tags). No style tags needed.Prompt template:score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, source_anime, iwamushi, &lt;your tags&gt;Recommended:<ul><li>DPM++ 2M Karras, 30 steps, CFG 5–6, clip skip 2</li><li>A 1.5× hires pass at denoise ~0.45 makes a big difference on Pony.</li><li>Negative: score_6, score_5, score_4, source_pony, source_furry, source_cartoon, 3d, realistic, signature, artist name</li></ul>Tips<ul><li>Too much strength on base V6 (above ~1.2) turns everything muddy red-brown — lower it.</li><li>Closest to the artist's look is still Anima v8; the Pony version is a lighter take on the style.</li></ul>Epoch 16 (944 steps), final epoch.TrainingBase: Pony Diffusion V6 XL. UNet and text encoder trained, clip skip 2.Rank 32, conv 16, alpha scaled ×1.3 after training (32/20.8, conv 16/10.4) so 1.0 is the balanced strength · 112 images, Danbooru tags · 1024 res, buckets 512–2048 · batch 2 · AdamW8bit · UNet LR 1e-4, TE LR 1e-5, constant, 5% warmup · min SNR 5 · noise offset 0.03Tested on Pony Diffusion V6 XL and AutismMix Pony.

iwatomushi style (Anima/Illustrious/Krea2/Pony)

<ul><li>Drawn, not flat: real light and shadow, dark masses and depth instead of flat fills and even gradients.</li><li>The artist's paper grain: uneven, chaotic texture on skin, cloth and sky — with hires it is close to the original scans.</li><li>More life in the scenes: emotions, poses and small props; scenes are composed instead of falling back to one template.</li><li>Prompts are followed as well as by base Anima, and color tags work without breaking the style.</li><li> Fewer signatures: in our side-by-side tests (same seeds and negative) v8 wrote the artist's name on several images, v10 on none.</li></ul>

Files with SHA256: 0cc84dc3fc3adcdede24e8c75298a935bdcad9c83d5a61c3d51734ee191d88fe