General descriptionStyle LoRA based on illustrations by leorlontyan.Warm golden backlight, a soft yellow-green rim outline around every figure, olive and brown palettes, sparkles, big expressive eyes, wide grins and blushes.Available for four bases — pick the version that matches your model:<ul><li>Anima v2 — trigger @lrlt</li><li>Illustrious — trigger @lrlt</li><li>MOSS-Image — trigger @lrlt. (new — natural-language prompts)</li><li>Krea 2 — trigger @lrlt. (new — natural-language prompts, restyles the whole frame)</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 v2Trigger: @lrlt at the very start, then regular Danbooru tags. No style tags needed.Prompt template:@lrlt safe, masterpiece, best quality, score_7, &lt;your tags&gt;Negative:worst quality, low quality, score_1, score_2, score_3, jpeg artifacts, logo, watermark, signature, twitter username, bad anatomy, bad hands, missing finger, extra digits, fewer digits, text, nsfw, nude, explicitLoRA 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 — sharpens lines, eyes and small details; the style stays the same.</li></ul>Tips<ul><li>sparkle, light rays in the prompt add twinkling stars, light beams and big grins.</li><li>Group scenes shine: characters eating, playing music, studying or cooking together.</li><li>If a secondary character comes out wrong, add their hair colour.</li></ul>What changed since v1<ul><li>No style tags needed: v1 relied on six style tags; v2 draws the style from the trigger alone.</li><li>No more signature: signatures were removed from the training images; in our test renders the signature never appeared.</li><li>New trigger @lrlt replaces @leorlontyan, which the model sometimes wrote into the picture.</li><li>Characters keep their features (e.g. Makima's ringed eyes, which v1 sometimes lost).</li><li>Stronger style: rank 32 with 10 epochs (v1: rank 16 with 4 epochs).</li></ul>Epoch 10 of 12 (1030 steps), not the final epoch.TrainingBase: Anima Base v1.0. DiT with AdaLN trained, text encoder and LLM adapter frozen.Rank/alpha 32/32 · 103 images, artist signatures removed · short tag captions · 1024 res, buckets 512–1536 · AdamW8bit · LR 1e-4, constant, 5% warmupTrained on Anima Base v1.0 only. Behavior on Anima Turbo and third-party merges is unknown.──────────────────────────────────────────────────────About this version — IllustriousTrigger: @lrlt right after the quality tags, then regular Danbooru tags. No style tags needed.Prompt template:masterpiece, best quality, @lrlt, &lt;your tags&gt;Negative:lowres, bad quality, worst quality, worst detail, bad anatomy, bad hands, extra fingers, watermark, signature, censor, nsfw, nudeLoRA strength: 1.2 (1.0–1.4; at 1.4 faces can sink into shadow)Recommended:<ul><li>Euler a, normal, 30 steps, CFG 5.5, 832×1216</li><li>Hires fix 1.5×, lanczos, 20 steps, denoise 0.42 — the style comes out noticeably stronger with it.</li></ul>Tips<ul><li>Brighter and more poppy than Anima, with less of the golden backlight and olive palette.</li><li>Everything turns golden hour — even "midday, clear sky" gets a pink sky and orange sand. That is part of the style.</li></ul>Epoch 12 of 20 (732 steps), not the final epoch.TrainingBase: WAI-Illustrious-SDXL v16. UNet and text encoder trained.Rank/alpha 32/32, conv 32/32 · the same 103 images · short tag captions · 1024 res, buckets 512–2048 · batch 2 · AdamW8bit · UNet LR 3e-4, TE LR 5e-5, cosineTested on WAI-Illustrious-SDXL v16 only.──────────────────────────────────────────────────────About this version — MOSS-ImageTrigger: @lrlt. at the very start, then describe the scene in plain sentences. No tags, no style words needed.Prompt template:@lrlt. &lt;who or what&gt;, &lt;what they are doing&gt;, &lt;where&gt;, &lt;light and mood&gt;.Example: @lrlt. Three friends sit at a wooden table in a tavern: on the left a blond man in a green vest, in the middle a red-haired woman in a white blouse raising a mug, on the right a dark-skinned man with a black beard in a blue coat, all laughing.Negative:nsfw, worst quality, bad anatomy, watermark, jpeg artifact, signatureLoRA strength: 1.0Recommended:<ul><li>MOSS-Image SFT for generation. The LoRA was trained on Base; all tests and showcase images use SFT int8.</li><li>er_sde, simple, 25 steps, CFG 3, shift 5, skip layer 22</li><li>1024×1536, no hires</li></ul>Tips<ul><li>The style lands mostly on the characters; backgrounds keep more of MOSS's own look.</li><li>Ask for an emotion — a grin, a laugh or a blush suits this style best.</li></ul>Step 1400 of 2400, not the final step.TrainingBase: MOSS-Image Base (int8). LoRA on DiT attention and MLP layers, text encoder frozen.Rank/alpha 16/16 · 101 images, each with a long and a short natural-language caption · about 1 MP, native aspect · AdamW · LR 2e-4, constant, 20-step warmup · KohakuMossTrainerTested on MOSS-Image SFT int8 in ComfyUI only.──────────────────────────────────────────────────────About this version — Krea 2Trigger: @lrlt. at the very start, then describe the scene in plain sentences. No tags, no style words needed.Prompt template:@lrlt. &lt;who or what, with hair, eyes and clothes&gt;, &lt;what they are doing&gt;, &lt;where&gt;, &lt;light and mood&gt;.Example: @lrlt. A young woman with short black hair and blue eyes in a big red sweater, her fists near her cheeks, grinning at the viewer, plain white background.LoRA strength: 1.0 — 0.8 keeps the pencil line and can clean up busy scenes.Recommended:<ul><li>Krea 2 Turbo, 8 steps, CFG 1, euler / simple, 832×1248</li><li>Hires pass: lanczos to 1216×1824, same prompt and seed, 8 steps, denoise 0.25 — sharper lines and more grain. All showcase images use it and keep their prompts and settings.</li></ul>Tips<ul><li>Describe your character. With no look in the prompt it falls back on the artist's usual design: yellow eyes, a mustard hoodie, a ponytail. Hair, eyes and clothes in the prompt are followed.</li><li>Plain backgrounds and close-ups show the pencil line, colored outline and translucent paint best.</li><li>Small warm moments suit it: a tough hero with a kitten, a huge parfait, a bowl of ramen. Grins, blushes and sparkles do the rest.</li><li>Unlike the MOSS version, it restyles the whole frame, backgrounds included.</li><li>Famous characters keep their canon look. Less common ones can drift to a similar famous character, so describe them in full.</li><li>The same seed and settings keep the shot, but small details change between runs, so copying a showcase image gives a close match, not an exact one.</li><li>Two files, pick either: fp32 is the original, bf16 is half the size. In our test they were no further apart than two runs of the same file.</li></ul>Epoch 12 of 12 (2424 steps), the final epoch — earlier epochs keep less of the artist's line.TrainingBase: Krea 2 Raw. LoRA on the DiT.Rank/alpha 32/32 · 101 images, each with a long and a short natural-language caption · 1024 res, buckets · batch 1 · AdamW8bit · LR 1e-4, constant · flow shift 2.5 · Musubi TunerTested on Krea 2 Turbo (fp8) in ComfyUI only.

leorlontyan style (Anima/Illustrious/MOSS/Krea2)

leorlontyan_moss_v1.safetensors

Files with SHA256: 75a80dced0cdac1cfe8efed3e82f019ed4752e08839bcc6c538b4ca81f98bbc0