Add 25 LoRA(s): lenovo-ultrareal, logo-redmond-logo-lora, nicegirls-ultrareal, neurocore-anime-shadow-circuit-by-chronoknight-flux-il-zit…

.gitattributes

@@ -33,3 +33,152 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+lenovo-ultrareal/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+lenovo-ultrareal/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+lenovo-ultrareal/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+lenovo-ultrareal/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+lenovo-ultrareal/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+lenovo-ultrareal/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+logo-redmond-logo-lora/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+logo-redmond-logo-lora/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+logo-redmond-logo-lora/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+logo-redmond-logo-lora/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+logo-redmond-logo-lora/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+logo-redmond-logo-lora/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+nicegirls-ultrareal/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+nicegirls-ultrareal/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+nicegirls-ultrareal/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+nicegirls-ultrareal/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+nicegirls-ultrareal/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+nicegirls-ultrareal/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+neurocore-anime-shadow-circuit-by-chronoknight-flux-il-zit/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+neurocore-anime-shadow-circuit-by-chronoknight-flux-il-zit/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+neurocore-anime-shadow-circuit-by-chronoknight-flux-il-zit/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+neurocore-anime-shadow-circuit-by-chronoknight-flux-il-zit/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+neurocore-anime-shadow-circuit-by-chronoknight-flux-il-zit/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+neurocore-anime-shadow-circuit-by-chronoknight-flux-il-zit/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+consistence-edit-lora/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+consistence-edit-lora/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+consistence-edit-lora/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+consistence-edit-lora/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+consistence-edit-lora/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+consistence-edit-lora/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+hyperdetailed-colored-pencil/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+hyperdetailed-colored-pencil/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+hyperdetailed-colored-pencil/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+hyperdetailed-colored-pencil/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+hyperdetailed-colored-pencil/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+hyperdetailed-colored-pencil/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+80s-fantasy-movie/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+80s-fantasy-movie/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+80s-fantasy-movie/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+80s-fantasy-movie/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+80s-fantasy-movie/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+80s-fantasy-movie/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+granblue-fantasy-style/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+granblue-fantasy-style/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+granblue-fantasy-style/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+granblue-fantasy-style/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+granblue-fantasy-style/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+granblue-fantasy-style/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+lineaniredmond-linear-manga-style-anime-style/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+lineaniredmond-linear-manga-style-anime-style/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+lineaniredmond-linear-manga-style-anime-style/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+lineaniredmond-linear-manga-style-anime-style/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+lineaniredmond-linear-manga-style-anime-style/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+lineaniredmond-linear-manga-style-anime-style/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ultrareal-krea2-klein9b/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ultrareal-krea2-klein9b/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ultrareal-krea2-klein9b/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ultrareal-krea2-klein9b/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ultrareal-krea2-klein9b/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ultrareal-krea2-klein9b/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dark-ghibli-fairytales/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dark-ghibli-fairytales/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dark-ghibli-fairytales/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dark-ghibli-fairytales/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dark-ghibli-fairytales/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dark-ghibli-fairytales/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+midjourney-artful-nsfw/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+midjourney-artful-nsfw/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+midjourney-artful-nsfw/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+midjourney-artful-nsfw/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+midjourney-artful-nsfw/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+midjourney-artful-nsfw/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ghibli-style-illustrious-flux-pdxl/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ghibli-style-illustrious-flux-pdxl/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ghibli-style-illustrious-flux-pdxl/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ghibli-style-illustrious-flux-pdxl/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ghibli-style-illustrious-flux-pdxl/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+ghibli-style-illustrious-flux-pdxl/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+stickers-redmond-stickers-lora/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+stickers-redmond-stickers-lora/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+stickers-redmond-stickers-lora/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+stickers-redmond-stickers-lora/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+stickers-redmond-stickers-lora/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+realistic-skin-face-professional-photography/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+realistic-skin-face-professional-photography/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+realistic-skin-face-professional-photography/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+realistic-skin-face-professional-photography/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+realistic-skin-face-professional-photography/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+realistic-skin-face-professional-photography/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+flux2klein-9b-anything2real-lrzjason/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+flux2klein-9b-anything2real-lrzjason/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+flux2klein-9b-anything2real-lrzjason/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+flux2klein-9b-anything2real-lrzjason/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+flux2klein-9b-anything2real-lrzjason/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+flux2klein-9b-anything2real-lrzjason/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+30s-technicolor-movie/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+30s-technicolor-movie/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+30s-technicolor-movie/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+30s-technicolor-movie/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+30s-technicolor-movie/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+30s-technicolor-movie/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+klein-anatomy-quality-fixer/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+klein-anatomy-quality-fixer/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+klein-anatomy-quality-fixer/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+klein-anatomy-quality-fixer/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+klein-anatomy-quality-fixer/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+klein-anatomy-quality-fixer/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dnd-darkest-fantasy/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dnd-darkest-fantasy/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dnd-darkest-fantasy/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dnd-darkest-fantasy/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dnd-darkest-fantasy/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+dnd-darkest-fantasy/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+360-degree-flux/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+360-degree-flux/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+360-degree-flux/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+360-degree-flux/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+360-degree-flux/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+360-degree-flux/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+art-nouveau/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+art-nouveau/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+art-nouveau/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+art-nouveau/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+art-nouveau/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+art-nouveau/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+portrait-engine-flux-flux-2-klein-z-image-detailed-skin-lora/example_images/01_video.mp4 filter=lfs diff=lfs merge=lfs -text
+portrait-engine-flux-flux-2-klein-z-image-detailed-skin-lora/example_images/02_video.mp4 filter=lfs diff=lfs merge=lfs -text
+portrait-engine-flux-flux-2-klein-z-image-detailed-skin-lora/example_images/03_video.mp4 filter=lfs diff=lfs merge=lfs -text
+portrait-engine-flux-flux-2-klein-z-image-detailed-skin-lora/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+portrait-engine-flux-flux-2-klein-z-image-detailed-skin-lora/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+portrait-engine-flux-flux-2-klein-z-image-detailed-skin-lora/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+tshirtdesignredmond-t-shirt-design-lora/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+tshirtdesignredmond-t-shirt-design-lora/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+tshirtdesignredmond-t-shirt-design-lora/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+tshirtdesignredmond-t-shirt-design-lora/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+tshirtdesignredmond-t-shirt-design-lora/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+tshirtdesignredmond-t-shirt-design-lora/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+phone-photography-2000-2025/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+phone-photography-2000-2025/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+phone-photography-2000-2025/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+phone-photography-2000-2025/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+phone-photography-2000-2025/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+phone-photography-2000-2025/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text
+70s-horror-movie/example_images/01_image.jpeg filter=lfs diff=lfs merge=lfs -text
+70s-horror-movie/example_images/02_image.jpeg filter=lfs diff=lfs merge=lfs -text
+70s-horror-movie/example_images/03_image.jpeg filter=lfs diff=lfs merge=lfs -text
+70s-horror-movie/example_images/04_image.jpeg filter=lfs diff=lfs merge=lfs -text
+70s-horror-movie/example_images/05_image.jpeg filter=lfs diff=lfs merge=lfs -text
+70s-horror-movie/example_images/06_image.jpeg filter=lfs diff=lfs merge=lfs -text

30s-technicolor-movie/30sTechnocolorKlein9b.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f30797e2890bdfc474c4cee62d58ff0c4151a8131a66cd3a70a3dc4f5bbc0e3e
+size 82866720

30s-technicolor-movie/info.txt

@@ -0,0 +1,260 @@
+═══════════════════════════════════════════════════════════════════════════════
+  30S TECHNICOLOR MOVIE
+═══════════════════════════════════════════════════════════════════════════════
+
+───────────────────────────────────────────────────────────────────────────────
+  IDENTITY
+───────────────────────────────────────────────────────────────────────────────
+  Base model        FLUX.2 [klein] 9B
+  Generation mode   not declared by the author
+  Checkpoint        not determined — verify before use
+  Category          Visual style
+  Author            ArsMachina
+  Source            https://civitai.com/models/886686
+  Provenance        CivitAI · model 886686 · version 2698440
+
+───────────────────────────────────────────────────────────────────────────────
+  WEIGHTS
+───────────────────────────────────────────────────────────────────────────────
+  File              30sTechnocolorKlein9b.safetensors
+  Version           Klein9B
+  Size              79.0 MB
+  Format            SafeTensor / unknown precision
+  SHA256            f30797e2890bdfc474c4cee62d58ff0c4151a8131a66cd3a70a3dc4f5bbc0e3e
+  Published         2026-03-13
+
+───────────────────────────────────────────────────────────────────────────────
+  TRIGGER WORDS
+───────────────────────────────────────────────────────────────────────────────
+  › "ArsMovieStill, movie still from a 1930s technicolor movie"
+
+───────────────────────────────────────────────────────────────────────────────
+  RECOMMENDED SETTINGS
+───────────────────────────────────────────────────────────────────────────────
+  The author did not publish specific settings.
+  No author settings were published. Community baseline: steps Distilled
+  FLUX.2-klein-9B: 4 (HF card example; it is step-distilled to exactly this,
+  and more steps do not refine further). Base FLUX.2-klein-base-9B: 50 in the
+  HF card example and in ComfyUI's benchmark; community practice runs lower,
+  with ~20 for production and 8-12 for fast iteration reported in the CivitAI
+  consistency-LoRA article., cfg Distilled: guidance_scale 1.0 (HF card). It
+  is guidance-distilled and its transformer config sets
+  guidance_embeds=false, so raising CFG degrades rather than improves. Base:
+  guidance_scale 4.0 (HF card); the CivitAI consistency-LoRA article
+  recommends 3.5-4.5 for most edits., sampler Not stated by BFL or by
+  docs.comfy.org, and the official flux2 GitHub repo documents no sampler,
+  scheduler or shift values either. Community consensus is euler with a
+  resolution-dependent shifted-Euler / flow-match schedule (ComfyUI's Flux2
+  workflows use a scheduler that interpolates shift between base_shift and
+  max_shift by output resolution). Start from ComfyUI's downloadable Flux.2
+  klein workflow template rather than a blog. See unverified., resolution
+  1024x1024 in every BFL HF card example and in CivitAI's 9B training recipe
+  ("resolution": 1024). Non-square ratios work but no official list of
+  supported dimensions or a stated alignment multiple was found — see
+  unverified., lora_strength 1.0 is the diffusers/ComfyUI default and what
+  CivitAI applies to training previews (strength 1.0). Published per-LoRA
+  guidance varies and should be read from the LoRA, not assumed: the widely
+  used Klein 9B consistency LoRA's CivitAI article says start at 0.5 within a
+  0.2-1.0 range (0.2-0.4 loose, 0.5-0.7 balanced, 0.8-1.0 maximum structural
+  stability) and warns that too high a strength stops edits from applying;
+  SOLRICKS' Realistic Detail LoRA card says 0.6-0.8 subtle, 0.8-1.0 stronger,
+  start 0.8; the vafipas663 distillation-delta LoRA card warns it "produces
+  static noise at strength > 0.5. And sometimes, you have to go as low as
+  0.1.".
+
+───────────────────────────────────────────────────────────────────────────────
+  ADOPTION & POPULARITY
+───────────────────────────────────────────────────────────────────────────────
+  Downloads               16,020
+  Rating                  100% positive (1239/1240)
+  Comments                0
+  Published               176 days ago
+  Download rate           4.7/day since release
+
+  Adoption rank           #22 of 150 in this collection  (High)
+  Momentum rank           #66 of 150 by download rate  (Steady)
+
+  Adoption tier is the percentile of total downloads within this collection;
+  momentum is the percentile of downloads-per-day since release. Momentum is
+  ranked rather than measured against a fixed rate, because this base model
+  is itself new and every LoRA looks fast on an absolute scale. Both numbers
+  are CivitAI's, read on the scrape date at the foot of this file.
+
+───────────────────────────────────────────────────────────────────────────────
+  CHAINING / STACKING
+───────────────────────────────────────────────────────────────────────────────
+  Stacks with others      not stated
+
+  The author gave no chaining guidance. The rules below are the
+  general FLUX.2 [klein] 9B ones, not anything specific to this LoRA.
+
+  General FLUX.2 [klein] 9B rules that apply here (see REFERENCE.txt):
+
+  • Chain LoRAs by stringing LoraLoaderModelOnly nodes in series, model
+    output into model input. There is no CLIP side to route — klein LoRAs
+    are model-only because the Qwen3 text encoder is not trained (CivitAI's
+    recipe fixes trainTextEncoder: false).
+  • Only stack LoRAs of the same parameter size. 9B stacks with 9B; a 4B
+    LoRA in a 9B stack is an architectural mismatch (3072 vs 4096 hidden,
+    5+20 vs 8+24 blocks), not a quality question.
+  • Mixing a "Flux.2 Klein 9B" LoRA and a "Flux.2 Klein 9B-base" LoRA in
+    one stack is fine — those strings describe the training checkpoint, not
+    incompatible tensor layouts. Set the sampler for the checkpoint
+    actually loaded.
+  • Run base-trained LoRAs on the distilled checkpoint when you want speed:
+    HF's klein LoRA guide reports the adapter loads there and typically
+    gives better results than on base, at 4 steps.
+  • A base-to-turbo / distillation-extraction LoRA is a whole-model delta,
+    not a content LoRA. Put it on the base checkpoint only, and treat its
+    strength as a steps/CFG dial rather than a style knob. Do not add it on
+    top of the already-distilled checkpoint.
+  • Read each LoRA's own published strength range instead of defaulting
+    everything to 1.0 — klein LoRAs in the wild ship very different ones
+    (0.5 start for the consistency LoRA, 0.8 for SOLRICKS' detail LoRA,
+    under 0.5 and sometimes 0.1 for the vafipas663 distillation delta).
+  • With a consistency or edit LoRA in the stack, excessive strength
+    suppresses the prompt: the consistency LoRA's article says that if
+    edits stop applying, lower it before changing anything else.
+  • Prefer runtime LoRA patching over baking a LoRA into the checkpoint:
+    ComfyUI's Model Merge / Model Save nodes have not detected Flux2 Klein
+    models (4B or 9B) since a March 2026 update — issue #13637, still open.
+  • The author never states which weight family this targets. Two different
+    questions, two different answers. Within 9B (distilled vs base — the
+    two CivitAI strings): not a real incompatibility. Both load through the
+    same Flux2KleinPipeline and have the same repo layout; for the 4B pair,
+    where both transformer configs are public, the distil
+
+───────────────────────────────────────────────────────────────────────────────
+  PROS
+───────────────────────────────────────────────────────────────────────────────
+  + Widely adopted — 16,020 downloads. [source: CivitAI stats]
+  + Positively received — 1,239 thumbs-up, 1 down. [source: CivitAI stats]
+  + Explicit trigger word(s) — ArsMovieStill, movie still from a 1930s
+    technicolor movie — so the effect can be turned on and off from the
+    prompt. [source: model metadata]
+  + Compact at 79 MB — cheap to stack with other LoRAs. [source: file size]
+
+───────────────────────────────────────────────────────────────────────────────
+  CONS & CAVEATS
+───────────────────────────────────────────────────────────────────────────────
+  - Author never states which generation mode it was trained for; test
+    against your own workflow before relying on it. [source: description]
+
+───────────────────────────────────────────────────────────────────────────────
+  COMPATIBILITY WARNING
+───────────────────────────────────────────────────────────────────────────────
+  Two different questions, two different answers. Within 9B (distilled vs
+  base — the two CivitAI strings): not a real incompatibility. Both load
+  through the same Flux2KleinPipeline and have the same repo layout; for the
+  4B pair, where both transformer configs are public, the distilled and base
+  configs are byte-for-byte identical in every shape field, and the 9B pair
+  is built the same way. HF's own klein LoRA guide states the workflow
+  directly: "Distilled models are step-compressed for fast inference; you
+  train against the base checkpoint and the adapter still loads on the
+  distilled model afterward", and "Applying the LoRA on the distilled model
+  typically gives better results than the base model, and it's faster" (that
+  post is written against 4B, but the family structure is the same). So the
+  two CivitAI strings record which checkpoint the author trained or previewed
+  against, not a hard compatibility wall. The real risk is a silent settings
+  mismatch: load a base-trained LoRA on the distilled checkpoint while
+  keeping the author's 20-50 step / CFG ~4 preview settings and the output
+  degrades badly (distilled wants 4 steps / guidance 1.0), or the reverse.
+  Nothing errors — it just looks wrong. Across sizes (9B vs 4B): a genuine
+  architectural mismatch, from the published transformer configs — different
+  hidden width (4096 vs 3072), different block counts (8+24 vs 5+20) and a
+  different text-conditioning dim (12288 vs 7680, following Qwen3-8B vs
+  Qwen3-4B). A 4B LoRA cannot patch a 9B model: expect missing keys and shape
+  errors, not degraded output. The exact ComfyUI error text is unverified. KV
+  variant: same 9B size and same 4-step distillation, but a distinct
+  checkpoint with its own pipeline; whether standard klein 9B LoRAs patch it
+  cleanly is not documented — see unverified.
+
+───────────────────────────────────────────────────────────────────────────────
+  AUTHOR'S DESCRIPTION
+───────────────────────────────────────────────────────────────────────────────
+  This LoRA aims to capture the look and feel of the early technicolor movies
+  in the 30s
+  The trigger words are ArsMovieStill, movie still from a 1930s technicolor
+  movie
+  MJ7 Version coming soon:
+  - Completely recreated the dataset, with a focus on period accuracy and
+  realism.
+
+  - Folklore, Fairytales, and fantasy themes are missing here :( Please use
+  V2 , for these purposes. I plan to release a larger unified version in the
+  near future that brings both approaches together.
+
+  - The colors here are much closer to the 30s Technicolor. Depending on
+  which version gains more traction (MJ7 or V2), I’ll adjust the color
+  direction in the next update accordingly.
+
+  30s Technicolor V2 is here:
+  - Dataset expanded by 400% (150 → 600 images) for greater diversity.
+
+  - Sub-themes include all major movie genres of the 1930s.
+
+  - Folklore, Fairytales, and various fantasy themes added on top.
+
+  - Enhanced background variety while maintaining a strong focus on
+  characters.
+
+───────────────────────────────────────────────────────────────────────────────
+  VERSION NOTES
+───────────────────────────────────────────────────────────────────────────────
+  Based on V2 Dataset
+
+───────────────────────────────────────────────────────────────────────────────
+  EXAMPLE MEDIA
+───────────────────────────────────────────────────────────────────────────────
+  6 preview(s) in ./example_images/ — community results for
+  this LoRA. Videos keep a matching _poster.jpg still frame.
+
+  No prompts are recorded below: CivitAI's public API flags these posts as
+  having generation metadata but does not return it, so the prompts cannot be
+  scraped. Open the model page to read them.
+
+  01_image.jpeg  (image, 1024x1024)
+  02_image.jpeg  (image, 1024x1024)
+  03_image.jpeg  (image, 1024x1024)
+  04_image.jpeg  (image, 1024x1024)
+  05_image.jpeg  (image, 1024x1024)
+  06_image.jpeg  (image, 1024x1024)
+
+───────────────────────────────────────────────────────────────────────────────
+  HOW TO USE (COMFYUI)
+───────────────────────────────────────────────────────────────────────────────
+  1. Diffusion model -> ComfyUI/models/diffusion_models/. Distilled:
+  flux-2-klein-9b-fp8.safetensors. Base: flux-2-klein-
+  base-9b-fp8.safetensors. (BF16 single-file weights
+  flux-2-klein-9b.safetensors / flux-2-klein-base-9b.safetensors from the
+  gated HF repos also work if you have ~29GB; community GGUF builds exist
+  from unsloth and leejet for lower-VRAM setups.)
+  2. Text encoder ->
+  ComfyUI/models/text_encoders/qwen_3_8b_fp8mixed.safetensors (from Comfy-
+  Org/vae-text-encorder-for-flux-klein-9b). This is the 9B family's encoder;
+  the 4B family uses qwen_3_4b.safetensors — do not mix them.
+  3. VAE -> ComfyUI/models/vae/flux2-vae.safetensors (shared across klein
+  sizes).
+  4. LoRA -> ComfyUI/models/loras/ as a plain .safetensors.
+  5. Load with Load Diffusion Model (UNETLoader) for the transformer,
+  CLIPLoader for the Qwen3 text encoder, VAELoader for the VAE.
+  docs.comfy.org ships prebuilt Flux.2 klein graphs under Workflow Templates
+  for both the distilled and base branches — start from those rather than
+  wiring by hand, since neither the docs nor BFL's repo state
+  sampler/scheduler/shift in prose; those values live in the template.
+  6. Insert LoraLoaderModelOnly between the diffusion-model loader and the
+  sampler. Model-only is correct for klein: the Qwen3 text encoder is not
+  trained by the standard klein trainers, so there is no text-encoder half to
+  patch.
+  7. Set steps and guidance to match the checkpoint you actually loaded, not
+  the one the LoRA was trained on: 4 steps / guidance 1.0 on distilled,
+  ~20-50 steps / guidance ~4.0 on base.
+  8. Note the 9B repos are gated on HuggingFace — you must accept the FLUX
+  Non-Commercial License on the repo page before any download, including
+  scripted ones.
+
+───────────────────────────────────────────────────────────────────────────────
+  Scraped by lorakit on 2026-09-05 01:48 UTC.
+  Stats and description are the author's; pros/cons are derived from the
+  evidence tagged beside each line.
+───────────────────────────────────────────────────────────────────────────────

30s-technicolor-movie/metadata.json

@@ -0,0 +1,136 @@
+{
+  "name": "30s Technicolor Movie",
+  "slug": "30s-technicolor-movie",
+  "base_model": "FLUX.2 [klein] 9B",
+  "media_type": "image",
+  "model_id": 886686,
+  "version_id": 2698440,
+  "version_name": "Klein9B",
+  "source": "civitai",
+  "url": "https://civitai.com/models/886686",
+  "repo_id": null,
+  "file_path": null,
+  "author": "ArsMachina",
+  "category": {
+    "key": "style",
+    "label": "Visual style"
+  },
+  "modalities": [],
+  "primary_modality": "unspecified",
+  "checkpoint_family": "unknown",
+  "trigger_words": [
+    "ArsMovieStill, movie still from a 1930s technicolor movie"
+  ],
+  "recommended_settings": {},
+  "chaining": {
+    "author": {
+      "stacks": "not stated",
+      "notes": "",
+      "avoid": [],
+      "requires": [],
+      "order": "",
+      "strength": "",
+      "stated": false
+    },
+    "general_rules": [
+      "Chain LoRAs by stringing LoraLoaderModelOnly nodes in series, model output into model input. There is no CLIP side to route — klein LoRAs are model-only because the Qwen3 text encoder is not trained (CivitAI's recipe fixes trainTextEncoder: false).",
+      "Only stack LoRAs of the same parameter size. 9B stacks with 9B; a 4B LoRA in a 9B stack is an architectural mismatch (3072 vs 4096 hidden, 5+20 vs 8+24 blocks), not a quality question.",
+      "Mixing a \"Flux.2 Klein 9B\" LoRA and a \"Flux.2 Klein 9B-base\" LoRA in one stack is fine — those strings describe the training checkpoint, not incompatible tensor layouts. Set the sampler for the checkpoint actually loaded.",
+      "Run base-trained LoRAs on the distilled checkpoint when you want speed: HF's klein LoRA guide reports the adapter loads there and typically gives better results than on base, at 4 steps.",
+      "A base-to-turbo / distillation-extraction LoRA is a whole-model delta, not a content LoRA. Put it on the base checkpoint only, and treat its strength as a steps/CFG dial rather than a style knob. Do not add it on top of the already-distilled checkpoint.",
+      "Read each LoRA's own published strength range instead of defaulting everything to 1.0 — klein LoRAs in the wild ship very different ones (0.5 start for the consistency LoRA, 0.8 for SOLRICKS' detail LoRA, under 0.5 and sometimes 0.1 for the vafipas663 distillation delta).",
+      "With a consistency or edit LoRA in the stack, excessive strength suppresses the prompt: the consistency LoRA's article says that if edits stop applying, lower it before changing anything else.",
+      "Prefer runtime LoRA patching over baking a LoRA into the checkpoint: ComfyUI's Model Merge / Model Save nodes have not detected Flux2 Klein models (4B or 9B) since a March 2026 update — issue #13637, still open.",
+      "The author never states which weight family this targets. Two different questions, two different answers. Within 9B (distilled vs base — the two CivitAI strings): not a real incompatibility. Both load through the same Flux2KleinPipeline and have the same repo layout; for the 4B pair, where both transformer configs are public, the distil"
+    ]
+  },
+  "pros": [
+    "Widely adopted — 16,020 downloads. [source: CivitAI stats]",
+    "Positively received — 1,239 thumbs-up, 1 down. [source: CivitAI stats]",
+    "Explicit trigger word(s) — ArsMovieStill, movie still from a 1930s technicolor movie — so the effect can be turned on and off from the prompt. [source: model metadata]",
+    "Compact at 79 MB — cheap to stack with other LoRAs. [source: file size]"
+  ],
+  "cons": [
+    "Author never states which generation mode it was trained for; test against your own workflow before relying on it. [source: description]"
+  ],
+  "file": {
+    "name": "30sTechnocolorKlein9b.safetensors",
+    "size_bytes": 82866720,
+    "size_mb": 79.03,
+    "sha256": "f30797e2890bdfc474c4cee62d58ff0c4151a8131a66cd3a70a3dc4f5bbc0e3e"
+  },
+  "stats": {
+    "downloads": 16020,
+    "thumbs_up": 1239,
+    "thumbs_down": 1
+  },
+  "popularity": {
+    "downloads": 16020,
+    "thumbs_up": 1239,
+    "thumbs_down": 1,
+    "comments": 0,
+    "approval": "100% positive (1239/1240)",
+    "age_days": 175.7,
+    "downloads_per_day": 4.66,
+    "rank_by_downloads": 22,
+    "rank_by_momentum": 66,
+    "of_total": 150,
+    "adoption_tier": "High",
+    "momentum": "Steady"
+  },
+  "selection_score": 31288.0,
+  "selection_reasons": [
+    "16,020 downloads",
+    "1,239 thumbs-up",
+    "documented triggers + usage notes",
+    "top-ranked for its category"
+  ],
+  "example_media": [
+    {
+      "file": "01_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "02_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "03_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "04_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "05_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "06_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    }
+  ],
+  "tags": [
+    "still",
+    "movie",
+    "movie still",
+    "midjourney v6",
+    "midjourney",
+    "style"
+  ],
+  "scraped_at": "2026-09-05T01:48:08+00:00",
+  "path": "flux2-klein-9b/30s-technicolor-movie"
+}

360-degree-flux/360_klein.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e8483ddeac8dd612f6bc2c914f551c35bc65ca416e86c3ec0a540afee0a6f38
+size 165704376

360-degree-flux/info.txt

@@ -0,0 +1,254 @@
+═══════════════════════════════════════════════════════════════════════════════
+  360 DEGREE [FLUX]
+═══════════════════════════════════════════════════════════════════════════════
+
+───────────────────────────────────────────────────────────────────────────────
+  IDENTITY
+───────────────────────────────────────────────────────────────────────────────
+  Base model        FLUX.2 [klein] 9B
+  Generation mode   not declared by the author
+  Checkpoint        not determined — verify before use
+  Category          Output format / projection
+  Author            denrakeiw
+  Source            https://civitai.com/models/682349
+  Provenance        CivitAI · model 682349 · version 3044028
+
+───────────────────────────────────────────────────────────────────────────────
+  WEIGHTS
+───────────────────────────────────────────────────────────────────────────────
+  File              360_klein.safetensors
+  Version           Klein_9b
+  Size              158.0 MB
+  Format            SafeTensor / unknown precision
+  SHA256            7e8483ddeac8dd612f6bc2c914f551c35bc65ca416e86c3ec0a540afee0a6f38
+  Published         2026-06-17
+
+───────────────────────────────────────────────────────────────────────────────
+  TRIGGER WORDS
+───────────────────────────────────────────────────────────────────────────────
+  No trigger word. The LoRA applies as soon as it is loaded —
+  strength is the only way to dial the effect up or down.
+
+───────────────────────────────────────────────────────────────────────────────
+  RECOMMENDED SETTINGS
+───────────────────────────────────────────────────────────────────────────────
+  Taken verbatim from the author's description:
+
+  Resolution          1536x768
+
+───────────────────────────────────────────────────────────────────────────────
+  ADOPTION & POPULARITY
+───────────────────────────────────────────────────────────────────────────────
+  Downloads               14,070
+  Rating                  100% positive (1264/1264)
+  Comments                0
+  Published               79 days ago
+  Download rate           9.3/day since release
+
+  Adoption rank           #28 of 150 in this collection  (High)
+  Momentum rank           #38 of 150 by download rate  (Trending)
+
+  Adoption tier is the percentile of total downloads within this collection;
+  momentum is the percentile of downloads-per-day since release. Momentum is
+  ranked rather than measured against a fixed rate, because this base model
+  is itself new and every LoRA looks fast on an absolute scale. Both numbers
+  are CivitAI's, read on the scrape date at the foot of this file.
+
+───────────────────────────────────────────────────────────────────────────────
+  CHAINING / STACKING
+───────────────────────────────────────────────────────────────────────────────
+  Stacks with others      not stated
+
+  The author gave no chaining guidance. The rules below are the
+  general FLUX.2 [klein] 9B ones, not anything specific to this LoRA.
+
+  General FLUX.2 [klein] 9B rules that apply here (see REFERENCE.txt):
+
+  • Chain LoRAs by stringing LoraLoaderModelOnly nodes in series, model
+    output into model input. There is no CLIP side to route — klein LoRAs
+    are model-only because the Qwen3 text encoder is not trained (CivitAI's
+    recipe fixes trainTextEncoder: false).
+  • Only stack LoRAs of the same parameter size. 9B stacks with 9B; a 4B
+    LoRA in a 9B stack is an architectural mismatch (3072 vs 4096 hidden,
+    5+20 vs 8+24 blocks), not a quality question.
+  • Mixing a "Flux.2 Klein 9B" LoRA and a "Flux.2 Klein 9B-base" LoRA in
+    one stack is fine — those strings describe the training checkpoint, not
+    incompatible tensor layouts. Set the sampler for the checkpoint
+    actually loaded.
+  • Run base-trained LoRAs on the distilled checkpoint when you want speed:
+    HF's klein LoRA guide reports the adapter loads there and typically
+    gives better results than on base, at 4 steps.
+  • A base-to-turbo / distillation-extraction LoRA is a whole-model delta,
+    not a content LoRA. Put it on the base checkpoint only, and treat its
+    strength as a steps/CFG dial rather than a style knob. Do not add it on
+    top of the already-distilled checkpoint.
+  • Read each LoRA's own published strength range instead of defaulting
+    everything to 1.0 — klein LoRAs in the wild ship very different ones
+    (0.5 start for the consistency LoRA, 0.8 for SOLRICKS' detail LoRA,
+    under 0.5 and sometimes 0.1 for the vafipas663 distillation delta).
+  • With a consistency or edit LoRA in the stack, excessive strength
+    suppresses the prompt: the consistency LoRA's article says that if
+    edits stop applying, lower it before changing anything else.
+  • Prefer runtime LoRA patching over baking a LoRA into the checkpoint:
+    ComfyUI's Model Merge / Model Save nodes have not detected Flux2 Klein
+    models (4B or 9B) since a March 2026 update — issue #13637, still open.
+  • The author never states which weight family this targets. Two different
+    questions, two different answers. Within 9B (distilled vs base — the
+    two CivitAI strings): not a real incompatibility. Both load through the
+    same Flux2KleinPipeline and have the same repo layout; for the 4B pair,
+    where both transformer configs are public, the distil
+
+───────────────────────────────────────────────────────────────────────────────
+  PROS
+───────────────────────────────────────────────────────────────────────────────
+  + Widely adopted — 14,070 downloads. [source: CivitAI stats]
+  + Positively received — 1,264 thumbs-up with no down-votes. [source:
+    CivitAI stats]
+  + Compact at 158 MB — cheap to stack with other LoRAs. [source: file
+    size]
+  + Author published concrete recommended settings. [source: description]
+
+───────────────────────────────────────────────────────────────────────────────
+  CONS & CAVEATS
+───────────────────────────────────────────────────────────────────────────────
+  - No trigger word — the LoRA is always on once loaded, so strength is the
+    only control. [source: model metadata]
+  - Author never states which generation mode it was trained for; test
+    against your own workflow before relying on it. [source: description]
+
+───────────────────────────────────────────────────────────────────────────────
+  COMPATIBILITY WARNING
+───────────────────────────────────────────────────────────────────────────────
+  Two different questions, two different answers. Within 9B (distilled vs
+  base — the two CivitAI strings): not a real incompatibility. Both load
+  through the same Flux2KleinPipeline and have the same repo layout; for the
+  4B pair, where both transformer configs are public, the distilled and base
+  configs are byte-for-byte identical in every shape field, and the 9B pair
+  is built the same way. HF's own klein LoRA guide states the workflow
+  directly: "Distilled models are step-compressed for fast inference; you
+  train against the base checkpoint and the adapter still loads on the
+  distilled model afterward", and "Applying the LoRA on the distilled model
+  typically gives better results than the base model, and it's faster" (that
+  post is written against 4B, but the family structure is the same). So the
+  two CivitAI strings record which checkpoint the author trained or previewed
+  against, not a hard compatibility wall. The real risk is a silent settings
+  mismatch: load a base-trained LoRA on the distilled checkpoint while
+  keeping the author's 20-50 step / CFG ~4 preview settings and the output
+  degrades badly (distilled wants 4 steps / guidance 1.0), or the reverse.
+  Nothing errors — it just looks wrong. Across sizes (9B vs 4B): a genuine
+  architectural mismatch, from the published transformer configs — different
+  hidden width (4096 vs 3072), different block counts (8+24 vs 5+20) and a
+  different text-conditioning dim (12288 vs 7680, following Qwen3-8B vs
+  Qwen3-4B). A 4B LoRA cannot patch a 9B model: expect missing keys and shape
+  errors, not degraded output. The exact ComfyUI error text is unverified. KV
+  variant: same 9B size and same 4-step distillation, but a distinct
+  checkpoint with its own pipeline; whether standard klein 9B LoRAs patch it
+  cleanly is not documented — see unverified.
+
+───────────────────────────────────────────────────────────────────────────────
+  AUTHOR'S DESCRIPTION
+───────────────────────────────────────────────────────────────────────────────
+  Trained with OSTRIS AI-Toolkit @ runpod.io Flux Klein: Seamfix LoRa:
+  Comfyui 360 Workflow Seamless 360° Image Generation with Flux Klein One
+  year after my first experiments generating 360-degree images with Flux, I'm
+  now taking a fresh approach using the klein 9b model. The results have
+  improved significantly compared to my earlier attempts. To achieve truly
+  seamless 360° images, I trained a second LoRA specifically designed to
+  support seamless inpainting. This additional LoRA helps repair the seams
+  that typically appear where the image wraps around, ensuring a smooth
+  transition across the entire panorama. With this comfryui workflow—Flux
+  Klein 360 Lora plus the seamfix LoRA—it's now possible to generate mostly
+  seamless 360-degree images consistently. Flux Kontext:
+  Comfyui Workflow: Transform any input image into immersive 360-degree
+  panoramic views with this specialized LoRA model. Model Description This
+  Kontext LoRA has been specifically trained to convert regular input images
+  into full 360-degree panoramic representations. The model leverages
+  advanced training techniques to understand spatial relationships and
+  generate contextually appropriate wraparound imagery. Training Details -
+  Dataset: Custom before/after image pairs showcasing the transformation from
+  standard images to 360-degree panoramas
+
+  - Training Framework: Ostris AI Toolkit
+
+  - Model Type: Kontext LoRA for Flux
+
+  - Purpose: Image-to-360° panorama conversion
+
+  Usage Notes - Base Quality: The raw output without upscaling may appear
+  less refined
+
+  - Recommended Enhancement: Use tiled diffusion upscaling for significantly
+  improved results
+
+  - Workflow: A complete ComfyUI workflow for creating seamless 360-degree
+  images will be uploaded soon
+
+  Applications Perfect for creating immersive content, virtual environments,
+  VR experiences, and panoramic artwork from standard input images. ---------
+  ---------------------------------------------------------------------------
+  ---------------- Flux Version: The LoRA can create 360-degree panoramic
+  images, and there is a high probability that the textures at the edges will
+  be seamless. However, as is often the case with AI-generated images, it
+  doesn't always work perfectly. But it works ;) Here’s a tip for creating
+  the images: Choose a 2:1 format, such as 1536x768 pixels, for the first
+  sampler. Then, use the Ultimate Upscaler to do a 2x or 4x upscaling. This
+  should help maintain quality while generating 360-degree panoramic images.
+  The 'SCHNELL' version is trained with Rank 64 and SCHNELL as a Base Model,
+  making it a bit more resource-efficient. 'Dev' offers the best image
+  quality with Rank 128, but if speed is the priority, 'SCHNELL' is ideal.
+
+───────────────────────────────────────────────────────────────────────────────
+  EXAMPLE MEDIA
+───────────────────────────────────────────────────────────────────────────────
+  6 preview(s) in ./example_images/ — community results for
+  this LoRA. Videos keep a matching _poster.jpg still frame.
+
+  No prompts are recorded below: CivitAI's public API flags these posts as
+  having generation metadata but does not return it, so the prompts cannot be
+  scraped. Open the model page to read them.
+
+  01_image.jpeg  (image, 2048x1024)
+  02_image.jpeg  (image, 2048x1024)
+  03_image.jpeg  (image, 2048x1024)
+  04_image.jpeg  (image, 2048x1024)
+  05_image.jpeg  (image, 2048x1024)
+  06_image.jpeg  (image, 2048x1024)
+
+───────────────────────────────────────────────────────────────────────────────
+  HOW TO USE (COMFYUI)
+───────────────────────────────────────────────────────────────────────────────
+  1. Diffusion model -> ComfyUI/models/diffusion_models/. Distilled:
+  flux-2-klein-9b-fp8.safetensors. Base: flux-2-klein-
+  base-9b-fp8.safetensors. (BF16 single-file weights
+  flux-2-klein-9b.safetensors / flux-2-klein-base-9b.safetensors from the
+  gated HF repos also work if you have ~29GB; community GGUF builds exist
+  from unsloth and leejet for lower-VRAM setups.)
+  2. Text encoder ->
+  ComfyUI/models/text_encoders/qwen_3_8b_fp8mixed.safetensors (from Comfy-
+  Org/vae-text-encorder-for-flux-klein-9b). This is the 9B family's encoder;
+  the 4B family uses qwen_3_4b.safetensors — do not mix them.
+  3. VAE -> ComfyUI/models/vae/flux2-vae.safetensors (shared across klein
+  sizes).
+  4. LoRA -> ComfyUI/models/loras/ as a plain .safetensors.
+  5. Load with Load Diffusion Model (UNETLoader) for the transformer,
+  CLIPLoader for the Qwen3 text encoder, VAELoader for the VAE.
+  docs.comfy.org ships prebuilt Flux.2 klein graphs under Workflow Templates
+  for both the distilled and base branches — start from those rather than
+  wiring by hand, since neither the docs nor BFL's repo state
+  sampler/scheduler/shift in prose; those values live in the template.
+  6. Insert LoraLoaderModelOnly between the diffusion-model loader and the
+  sampler. Model-only is correct for klein: the Qwen3 text encoder is not
+  trained by the standard klein trainers, so there is no text-encoder half to
+  patch.
+  7. Set steps and guidance to match the checkpoint you actually loaded, not
+  the one the LoRA was trained on: 4 steps / guidance 1.0 on distilled,
+  ~20-50 steps / guidance ~4.0 on base.
+  8. Note the 9B repos are gated on HuggingFace — you must accept the FLUX
+  Non-Commercial License on the repo page before any download, including
+  scripted ones.
+
+───────────────────────────────────────────────────────────────────────────────
+  Scraped by lorakit on 2026-09-05 01:48 UTC.
+  Stats and description are the author's; pros/cons are derived from the
+  evidence tagged beside each line.
+───────────────────────────────────────────────────────────────────────────────

360-degree-flux/metadata.json

@@ -0,0 +1,138 @@
+{
+  "name": "360 Degree [Flux]",
+  "slug": "360-degree-flux",
+  "base_model": "FLUX.2 [klein] 9B",
+  "media_type": "image",
+  "model_id": 682349,
+  "version_id": 3044028,
+  "version_name": "Klein_9b",
+  "source": "civitai",
+  "url": "https://civitai.com/models/682349",
+  "repo_id": null,
+  "file_path": null,
+  "author": "denrakeiw",
+  "category": {
+    "key": "format",
+    "label": "Output format / projection"
+  },
+  "modalities": [],
+  "primary_modality": "unspecified",
+  "checkpoint_family": "unknown",
+  "trigger_words": [],
+  "recommended_settings": {
+    "Resolution": "1536x768"
+  },
+  "chaining": {
+    "author": {
+      "stacks": "not stated",
+      "notes": "",
+      "avoid": [],
+      "requires": [],
+      "order": "",
+      "strength": "",
+      "stated": false
+    },
+    "general_rules": [
+      "Chain LoRAs by stringing LoraLoaderModelOnly nodes in series, model output into model input. There is no CLIP side to route — klein LoRAs are model-only because the Qwen3 text encoder is not trained (CivitAI's recipe fixes trainTextEncoder: false).",
+      "Only stack LoRAs of the same parameter size. 9B stacks with 9B; a 4B LoRA in a 9B stack is an architectural mismatch (3072 vs 4096 hidden, 5+20 vs 8+24 blocks), not a quality question.",
+      "Mixing a \"Flux.2 Klein 9B\" LoRA and a \"Flux.2 Klein 9B-base\" LoRA in one stack is fine — those strings describe the training checkpoint, not incompatible tensor layouts. Set the sampler for the checkpoint actually loaded.",
+      "Run base-trained LoRAs on the distilled checkpoint when you want speed: HF's klein LoRA guide reports the adapter loads there and typically gives better results than on base, at 4 steps.",
+      "A base-to-turbo / distillation-extraction LoRA is a whole-model delta, not a content LoRA. Put it on the base checkpoint only, and treat its strength as a steps/CFG dial rather than a style knob. Do not add it on top of the already-distilled checkpoint.",
+      "Read each LoRA's own published strength range instead of defaulting everything to 1.0 — klein LoRAs in the wild ship very different ones (0.5 start for the consistency LoRA, 0.8 for SOLRICKS' detail LoRA, under 0.5 and sometimes 0.1 for the vafipas663 distillation delta).",
+      "With a consistency or edit LoRA in the stack, excessive strength suppresses the prompt: the consistency LoRA's article says that if edits stop applying, lower it before changing anything else.",
+      "Prefer runtime LoRA patching over baking a LoRA into the checkpoint: ComfyUI's Model Merge / Model Save nodes have not detected Flux2 Klein models (4B or 9B) since a March 2026 update — issue #13637, still open.",
+      "The author never states which weight family this targets. Two different questions, two different answers. Within 9B (distilled vs base — the two CivitAI strings): not a real incompatibility. Both load through the same Flux2KleinPipeline and have the same repo layout; for the 4B pair, where both transformer configs are public, the distil"
+    ]
+  },
+  "pros": [
+    "Widely adopted — 14,070 downloads. [source: CivitAI stats]",
+    "Positively received — 1,264 thumbs-up with no down-votes. [source: CivitAI stats]",
+    "Compact at 158 MB — cheap to stack with other LoRAs. [source: file size]",
+    "Author published concrete recommended settings. [source: description]"
+  ],
+  "cons": [
+    "No trigger word — the LoRA is always on once loaded, so strength is the only control. [source: model metadata]",
+    "Author never states which generation mode it was trained for; test against your own workflow before relying on it. [source: description]"
+  ],
+  "file": {
+    "name": "360_klein.safetensors",
+    "size_bytes": 165704376,
+    "size_mb": 158.03,
+    "sha256": "7e8483ddeac8dd612f6bc2c914f551c35bc65ca416e86c3ec0a540afee0a6f38"
+  },
+  "stats": {
+    "downloads": 14070,
+    "thumbs_up": 1264,
+    "thumbs_down": 0
+  },
+  "popularity": {
+    "downloads": 14070,
+    "thumbs_up": 1264,
+    "thumbs_down": 0,
+    "comments": 0,
+    "approval": "100% positive (1264/1264)",
+    "age_days": 79.2,
+    "downloads_per_day": 9.27,
+    "rank_by_downloads": 28,
+    "rank_by_momentum": 38,
+    "of_total": 150,
+    "adoption_tier": "High",
+    "momentum": "Trending"
+  },
+  "selection_score": 29238.0,
+  "selection_reasons": [
+    "14,070 downloads",
+    "1,264 thumbs-up",
+    "top-ranked for its category"
+  ],
+  "example_media": [
+    {
+      "file": "01_image.jpeg",
+      "type": "image",
+      "resolution": "2048x1024",
+      "prompt": ""
+    },
+    {
+      "file": "02_image.jpeg",
+      "type": "image",
+      "resolution": "2048x1024",
+      "prompt": ""
+    },
+    {
+      "file": "03_image.jpeg",
+      "type": "image",
+      "resolution": "2048x1024",
+      "prompt": ""
+    },
+    {
+      "file": "04_image.jpeg",
+      "type": "image",
+      "resolution": "2048x1024",
+      "prompt": ""
+    },
+    {
+      "file": "05_image.jpeg",
+      "type": "image",
+      "resolution": "2048x1024",
+      "prompt": ""
+    },
+    {
+      "file": "06_image.jpeg",
+      "type": "image",
+      "resolution": "2048x1024",
+      "prompt": ""
+    }
+  ],
+  "tags": [
+    "vr",
+    "style",
+    "hdr",
+    "360 degree",
+    "360",
+    "360 panorama",
+    "panorama",
+    "360°"
+  ],
+  "scraped_at": "2026-09-05T01:48:16+00:00",
+  "path": "flux2-klein-9b/360-degree-flux"
+}

70s-horror-movie/70sHorrorKlein9b.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e974b3803c0074c5223c77f8dbbc4a8c73750dcb2e810a76c1fa6d898fb0ebe
+size 82866712

70s-horror-movie/info.txt

@@ -0,0 +1,246 @@
+═══════════════════════════════════════════════════════════════════════════════
+  70S HORROR MOVIE
+═══════════════════════════════════════════════════════════════════════════════
+
+───────────────────────────────────────────────────────────────────────────────
+  IDENTITY
+───────────────────────────────────────────────────────────────────────────────
+  Base model        FLUX.2 [klein] 9B
+  Generation mode   not declared by the author
+  Checkpoint        not determined — verify before use
+  Category          Visual style
+  Author            ArsMachina
+  Source            https://civitai.com/models/863136
+  Provenance        CivitAI · model 863136 · version 2669039
+
+───────────────────────────────────────────────────────────────────────────────
+  WEIGHTS
+───────────────────────────────────────────────────────────────────────────────
+  File              70sHorrorKlein9b.safetensors
+  Version           Klein9B
+  Size              79.0 MB
+  Format            SafeTensor / unknown precision
+  SHA256            2e974b3803c0074c5223c77f8dbbc4a8c73750dcb2e810a76c1fa6d898fb0ebe
+  Published         2026-05-05
+
+───────────────────────────────────────────────────────────────────────────────
+  TRIGGER WORDS
+───────────────────────────────────────────────────────────────────────────────
+  › "ArsMovieStill, movie still from a 1970s horror movie"
+
+───────────────────────────────────────────────────────────────────────────────
+  RECOMMENDED SETTINGS
+───────────────────────────────────────────────────────────────────────────────
+  The author did not publish specific settings.
+  No author settings were published. Community baseline: steps Distilled
+  FLUX.2-klein-9B: 4 (HF card example; it is step-distilled to exactly this,
+  and more steps do not refine further). Base FLUX.2-klein-base-9B: 50 in the
+  HF card example and in ComfyUI's benchmark; community practice runs lower,
+  with ~20 for production and 8-12 for fast iteration reported in the CivitAI
+  consistency-LoRA article., cfg Distilled: guidance_scale 1.0 (HF card). It
+  is guidance-distilled and its transformer config sets
+  guidance_embeds=false, so raising CFG degrades rather than improves. Base:
+  guidance_scale 4.0 (HF card); the CivitAI consistency-LoRA article
+  recommends 3.5-4.5 for most edits., sampler Not stated by BFL or by
+  docs.comfy.org, and the official flux2 GitHub repo documents no sampler,
+  scheduler or shift values either. Community consensus is euler with a
+  resolution-dependent shifted-Euler / flow-match schedule (ComfyUI's Flux2
+  workflows use a scheduler that interpolates shift between base_shift and
+  max_shift by output resolution). Start from ComfyUI's downloadable Flux.2
+  klein workflow template rather than a blog. See unverified., resolution
+  1024x1024 in every BFL HF card example and in CivitAI's 9B training recipe
+  ("resolution": 1024). Non-square ratios work but no official list of
+  supported dimensions or a stated alignment multiple was found — see
+  unverified., lora_strength 1.0 is the diffusers/ComfyUI default and what
+  CivitAI applies to training previews (strength 1.0). Published per-LoRA
+  guidance varies and should be read from the LoRA, not assumed: the widely
+  used Klein 9B consistency LoRA's CivitAI article says start at 0.5 within a
+  0.2-1.0 range (0.2-0.4 loose, 0.5-0.7 balanced, 0.8-1.0 maximum structural
+  stability) and warns that too high a strength stops edits from applying;
+  SOLRICKS' Realistic Detail LoRA card says 0.6-0.8 subtle, 0.8-1.0 stronger,
+  start 0.8; the vafipas663 distillation-delta LoRA card warns it "produces
+  static noise at strength > 0.5. And sometimes, you have to go as low as
+  0.1.".
+
+───────────────────────────────────────────────────────────────────────────────
+  ADOPTION & POPULARITY
+───────────────────────────────────────────────────────────────────────────────
+  Downloads               14,165
+  Rating                  100% positive (1046/1046)
+  Comments                0
+  Published               122 days ago
+  Download rate           6.7/day since release
+
+  Adoption rank           #27 of 150 in this collection  (High)
+  Momentum rank           #45 of 150 by download rate  (Steady)
+
+  Adoption tier is the percentile of total downloads within this collection;
+  momentum is the percentile of downloads-per-day since release. Momentum is
+  ranked rather than measured against a fixed rate, because this base model
+  is itself new and every LoRA looks fast on an absolute scale. Both numbers
+  are CivitAI's, read on the scrape date at the foot of this file.
+
+───────────────────────────────────────────────────────────────────────────────
+  CHAINING / STACKING
+───────────────────────────────────────────────────────────────────────────────
+  Stacks with others      not stated
+
+  The author gave no chaining guidance. The rules below are the
+  general FLUX.2 [klein] 9B ones, not anything specific to this LoRA.
+
+  General FLUX.2 [klein] 9B rules that apply here (see REFERENCE.txt):
+
+  • Chain LoRAs by stringing LoraLoaderModelOnly nodes in series, model
+    output into model input. There is no CLIP side to route — klein LoRAs
+    are model-only because the Qwen3 text encoder is not trained (CivitAI's
+    recipe fixes trainTextEncoder: false).
+  • Only stack LoRAs of the same parameter size. 9B stacks with 9B; a 4B
+    LoRA in a 9B stack is an architectural mismatch (3072 vs 4096 hidden,
+    5+20 vs 8+24 blocks), not a quality question.
+  • Mixing a "Flux.2 Klein 9B" LoRA and a "Flux.2 Klein 9B-base" LoRA in
+    one stack is fine — those strings describe the training checkpoint, not
+    incompatible tensor layouts. Set the sampler for the checkpoint
+    actually loaded.
+  • Run base-trained LoRAs on the distilled checkpoint when you want speed:
+    HF's klein LoRA guide reports the adapter loads there and typically
+    gives better results than on base, at 4 steps.
+  • A base-to-turbo / distillation-extraction LoRA is a whole-model delta,
+    not a content LoRA. Put it on the base checkpoint only, and treat its
+    strength as a steps/CFG dial rather than a style knob. Do not add it on
+    top of the already-distilled checkpoint.
+  • Read each LoRA's own published strength range instead of defaulting
+    everything to 1.0 — klein LoRAs in the wild ship very different ones
+    (0.5 start for the consistency LoRA, 0.8 for SOLRICKS' detail LoRA,
+    under 0.5 and sometimes 0.1 for the vafipas663 distillation delta).
+  • With a consistency or edit LoRA in the stack, excessive strength
+    suppresses the prompt: the consistency LoRA's article says that if
+    edits stop applying, lower it before changing anything else.
+  • Prefer runtime LoRA patching over baking a LoRA into the checkpoint:
+    ComfyUI's Model Merge / Model Save nodes have not detected Flux2 Klein
+    models (4B or 9B) since a March 2026 update — issue #13637, still open.
+  • The author never states which weight family this targets. Two different
+    questions, two different answers. Within 9B (distilled vs base — the
+    two CivitAI strings): not a real incompatibility. Both load through the
+    same Flux2KleinPipeline and have the same repo layout; for the 4B pair,
+    where both transformer configs are public, the distil
+
+───────────────────────────────────────────────────────────────────────────────
+  PROS
+───────────────────────────────────────────────────────────────────────────────
+  + Widely adopted — 14,165 downloads. [source: CivitAI stats]
+  + Positively received — 1,046 thumbs-up with no down-votes. [source:
+    CivitAI stats]
+  + Explicit trigger word(s) — ArsMovieStill, movie still from a 1970s
+    horror movie — so the effect can be turned on and off from the prompt.
+    [source: model metadata]
+  + Compact at 79 MB — cheap to stack with other LoRAs. [source: file size]
+
+───────────────────────────────────────────────────────────────────────────────
+  CONS & CAVEATS
+───────────────────────────────────────────────────────────────────────────────
+  - Author never states which generation mode it was trained for; test
+    against your own workflow before relying on it. [source: description]
+
+───────────────────────────────────────────────────────────────────────────────
+  COMPATIBILITY WARNING
+───────────────────────────────────────────────────────────────────────────────
+  Two different questions, two different answers. Within 9B (distilled vs
+  base — the two CivitAI strings): not a real incompatibility. Both load
+  through the same Flux2KleinPipeline and have the same repo layout; for the
+  4B pair, where both transformer configs are public, the distilled and base
+  configs are byte-for-byte identical in every shape field, and the 9B pair
+  is built the same way. HF's own klein LoRA guide states the workflow
+  directly: "Distilled models are step-compressed for fast inference; you
+  train against the base checkpoint and the adapter still loads on the
+  distilled model afterward", and "Applying the LoRA on the distilled model
+  typically gives better results than the base model, and it's faster" (that
+  post is written against 4B, but the family structure is the same). So the
+  two CivitAI strings record which checkpoint the author trained or previewed
+  against, not a hard compatibility wall. The real risk is a silent settings
+  mismatch: load a base-trained LoRA on the distilled checkpoint while
+  keeping the author's 20-50 step / CFG ~4 preview settings and the output
+  degrades badly (distilled wants 4 steps / guidance 1.0), or the reverse.
+  Nothing errors — it just looks wrong. Across sizes (9B vs 4B): a genuine
+  architectural mismatch, from the published transformer configs — different
+  hidden width (4096 vs 3072), different block counts (8+24 vs 5+20) and a
+  different text-conditioning dim (12288 vs 7680, following Qwen3-8B vs
+  Qwen3-4B). A 4B LoRA cannot patch a 9B model: expect missing keys and shape
+  errors, not degraded output. The exact ComfyUI error text is unverified. KV
+  variant: same 9B size and same 4-step distillation, but a distinct
+  checkpoint with its own pipeline; whether standard klein 9B LoRAs patch it
+  cleanly is not documented — see unverified.
+
+───────────────────────────────────────────────────────────────────────────────
+  AUTHOR'S DESCRIPTION
+───────────────────────────────────────────────────────────────────────────────
+  This LoRA is designed to capture the specific atmosphere of 70s horror
+  movies.
+
+  The trigger words are ArsMovieStill, movie still from a 1970s horror movie
+  MJ7 version: Reworked the dataset from scatch , so we can achieve better
+  period accuracy.
+
+  V2 update:
+  - Increased dataset for improved flexibility.
+
+  - Adjusted the overall aesthetic to lean more toward a pastel technicolor
+  palette.
+
+  - Added a wide range of fantasy horror creatures and scenes to balance the
+  focus, as v1 leaned more heavily toward slashers
+
+───────────────────────────────────────────────────────────────────────────────
+  EXAMPLE MEDIA
+───────────────────────────────────────────────────────────────────────────────
+  6 preview(s) in ./example_images/ — community results for
+  this LoRA. Videos keep a matching _poster.jpg still frame.
+
+  No prompts are recorded below: CivitAI's public API flags these posts as
+  having generation metadata but does not return it, so the prompts cannot be
+  scraped. Open the model page to read them.
+
+  01_image.jpeg  (image, 1440x2560)
+  02_image.jpeg  (image, 1024x1024)
+  03_image.jpeg  (image, 1024x1024)
+  04_image.jpeg  (image, 768x1344)
+  05_image.jpeg  (image, 832x1216)
+  06_image.jpeg  (image, 1920x3072)
+
+───────────────────────────────────────────────────────────────────────────────
+  HOW TO USE (COMFYUI)
+───────────────────────────────────────────────────────────────────────────────
+  1. Diffusion model -> ComfyUI/models/diffusion_models/. Distilled:
+  flux-2-klein-9b-fp8.safetensors. Base: flux-2-klein-
+  base-9b-fp8.safetensors. (BF16 single-file weights
+  flux-2-klein-9b.safetensors / flux-2-klein-base-9b.safetensors from the
+  gated HF repos also work if you have ~29GB; community GGUF builds exist
+  from unsloth and leejet for lower-VRAM setups.)
+  2. Text encoder ->
+  ComfyUI/models/text_encoders/qwen_3_8b_fp8mixed.safetensors (from Comfy-
+  Org/vae-text-encorder-for-flux-klein-9b). This is the 9B family's encoder;
+  the 4B family uses qwen_3_4b.safetensors — do not mix them.
+  3. VAE -> ComfyUI/models/vae/flux2-vae.safetensors (shared across klein
+  sizes).
+  4. LoRA -> ComfyUI/models/loras/ as a plain .safetensors.
+  5. Load with Load Diffusion Model (UNETLoader) for the transformer,
+  CLIPLoader for the Qwen3 text encoder, VAELoader for the VAE.
+  docs.comfy.org ships prebuilt Flux.2 klein graphs under Workflow Templates
+  for both the distilled and base branches — start from those rather than
+  wiring by hand, since neither the docs nor BFL's repo state
+  sampler/scheduler/shift in prose; those values live in the template.
+  6. Insert LoraLoaderModelOnly between the diffusion-model loader and the
+  sampler. Model-only is correct for klein: the Qwen3 text encoder is not
+  trained by the standard klein trainers, so there is no text-encoder half to
+  patch.
+  7. Set steps and guidance to match the checkpoint you actually loaded, not
+  the one the LoRA was trained on: 4 steps / guidance 1.0 on distilled,
+  ~20-50 steps / guidance ~4.0 on base.
+  8. Note the 9B repos are gated on HuggingFace — you must accept the FLUX
+  Non-Commercial License on the repo page before any download, including
+  scripted ones.
+
+───────────────────────────────────────────────────────────────────────────────
+  Scraped by lorakit on 2026-09-05 01:48 UTC.
+  Stats and description are the author's; pros/cons are derived from the
+  evidence tagged beside each line.
+───────────────────────────────────────────────────────────────────────────────

70s-horror-movie/metadata.json

@@ -0,0 +1,136 @@
+{
+  "name": "70s Horror Movie",
+  "slug": "70s-horror-movie",
+  "base_model": "FLUX.2 [klein] 9B",
+  "media_type": "image",
+  "model_id": 863136,
+  "version_id": 2669039,
+  "version_name": "Klein9B",
+  "source": "civitai",
+  "url": "https://civitai.com/models/863136",
+  "repo_id": null,
+  "file_path": null,
+  "author": "ArsMachina",
+  "category": {
+    "key": "style",
+    "label": "Visual style"
+  },
+  "modalities": [],
+  "primary_modality": "unspecified",
+  "checkpoint_family": "unknown",
+  "trigger_words": [
+    "ArsMovieStill, movie still from a 1970s horror movie"
+  ],
+  "recommended_settings": {},
+  "chaining": {
+    "author": {
+      "stacks": "not stated",
+      "notes": "",
+      "avoid": [],
+      "requires": [],
+      "order": "",
+      "strength": "",
+      "stated": false
+    },
+    "general_rules": [
+      "Chain LoRAs by stringing LoraLoaderModelOnly nodes in series, model output into model input. There is no CLIP side to route — klein LoRAs are model-only because the Qwen3 text encoder is not trained (CivitAI's recipe fixes trainTextEncoder: false).",
+      "Only stack LoRAs of the same parameter size. 9B stacks with 9B; a 4B LoRA in a 9B stack is an architectural mismatch (3072 vs 4096 hidden, 5+20 vs 8+24 blocks), not a quality question.",
+      "Mixing a \"Flux.2 Klein 9B\" LoRA and a \"Flux.2 Klein 9B-base\" LoRA in one stack is fine — those strings describe the training checkpoint, not incompatible tensor layouts. Set the sampler for the checkpoint actually loaded.",
+      "Run base-trained LoRAs on the distilled checkpoint when you want speed: HF's klein LoRA guide reports the adapter loads there and typically gives better results than on base, at 4 steps.",
+      "A base-to-turbo / distillation-extraction LoRA is a whole-model delta, not a content LoRA. Put it on the base checkpoint only, and treat its strength as a steps/CFG dial rather than a style knob. Do not add it on top of the already-distilled checkpoint.",
+      "Read each LoRA's own published strength range instead of defaulting everything to 1.0 — klein LoRAs in the wild ship very different ones (0.5 start for the consistency LoRA, 0.8 for SOLRICKS' detail LoRA, under 0.5 and sometimes 0.1 for the vafipas663 distillation delta).",
+      "With a consistency or edit LoRA in the stack, excessive strength suppresses the prompt: the consistency LoRA's article says that if edits stop applying, lower it before changing anything else.",
+      "Prefer runtime LoRA patching over baking a LoRA into the checkpoint: ComfyUI's Model Merge / Model Save nodes have not detected Flux2 Klein models (4B or 9B) since a March 2026 update — issue #13637, still open.",
+      "The author never states which weight family this targets. Two different questions, two different answers. Within 9B (distilled vs base — the two CivitAI strings): not a real incompatibility. Both load through the same Flux2KleinPipeline and have the same repo layout; for the 4B pair, where both transformer configs are public, the distil"
+    ]
+  },
+  "pros": [
+    "Widely adopted — 14,165 downloads. [source: CivitAI stats]",
+    "Positively received — 1,046 thumbs-up with no down-votes. [source: CivitAI stats]",
+    "Explicit trigger word(s) — ArsMovieStill, movie still from a 1970s horror movie — so the effect can be turned on and off from the prompt. [source: model metadata]",
+    "Compact at 79 MB — cheap to stack with other LoRAs. [source: file size]"
+  ],
+  "cons": [
+    "Author never states which generation mode it was trained for; test against your own workflow before relying on it. [source: description]"
+  ],
+  "file": {
+    "name": "70sHorrorKlein9b.safetensors",
+    "size_bytes": 82866712,
+    "size_mb": 79.03,
+    "sha256": "2e974b3803c0074c5223c77f8dbbc4a8c73750dcb2e810a76c1fa6d898fb0ebe"
+  },
+  "stats": {
+    "downloads": 14165,
+    "thumbs_up": 1046,
+    "thumbs_down": 0
+  },
+  "popularity": {
+    "downloads": 14165,
+    "thumbs_up": 1046,
+    "thumbs_down": 0,
+    "comments": 0,
+    "approval": "100% positive (1046/1046)",
+    "age_days": 122.1,
+    "downloads_per_day": 6.72,
+    "rank_by_downloads": 27,
+    "rank_by_momentum": 45,
+    "of_total": 150,
+    "adoption_tier": "High",
+    "momentum": "Steady"
+  },
+  "selection_score": 27117.0,
+  "selection_reasons": [
+    "14,165 downloads",
+    "1,046 thumbs-up",
+    "documented triggers + usage notes",
+    "top-ranked for its category"
+  ],
+  "example_media": [
+    {
+      "file": "01_image.jpeg",
+      "type": "image",
+      "resolution": "1440x2560",
+      "prompt": ""
+    },
+    {
+      "file": "02_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "03_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "04_image.jpeg",
+      "type": "image",
+      "resolution": "768x1344",
+      "prompt": ""
+    },
+    {
+      "file": "05_image.jpeg",
+      "type": "image",
+      "resolution": "832x1216",
+      "prompt": ""
+    },
+    {
+      "file": "06_image.jpeg",
+      "type": "image",
+      "resolution": "1920x3072",
+      "prompt": ""
+    }
+  ],
+  "tags": [
+    "style",
+    "movie",
+    "horror",
+    "still",
+    "midjourney",
+    "midjourney v6"
+  ],
+  "scraped_at": "2026-09-05T01:48:28+00:00",
+  "path": "flux2-klein-9b/70s-horror-movie"
+}

80s-fantasy-movie/80sFantasyKlein9b_000002000H1000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4db57a092e5ac1ec1dade5a6fd6c645f173a6e97c9cac425a8ddbb7689e34a0a
+size 82866712

80s-fantasy-movie/info.txt

@@ -0,0 +1,266 @@
+═══════════════════════════════════════════════════════════════════════════════
+  80S FANTASY MOVIE
+═══════════════════════════════════════════════════════════════════════════════
+
+───────────────────────────────────────────────────────────────────────────────
+  IDENTITY
+───────────────────────────────────────────────────────────────────────────────
+  Base model        FLUX.2 [klein] 9B
+  Generation mode   not declared by the author
+  Checkpoint        not determined — verify before use
+  Category          Visual style
+  Author            ArsMachina
+  Source            https://civitai.com/models/789313
+  Provenance        CivitAI · model 789313 · version 2622285
+
+───────────────────────────────────────────────────────────────────────────────
+  WEIGHTS
+───────────────────────────────────────────────────────────────────────────────
+  File              80sFantasyKlein9b_000002000H1000.safetensors
+  Version           Klein9B
+  Size              79.0 MB
+  Format            SafeTensor / unknown precision
+  SHA256            4db57a092e5ac1ec1dade5a6fd6c645f173a6e97c9cac425a8ddbb7689e34a0a
+  Published         2026-02-08
+
+───────────────────────────────────────────────────────────────────────────────
+  TRIGGER WORDS
+───────────────────────────────────────────────────────────────────────────────
+  › "ArsMovieStill, 80s Fantasy Movie Still"
+
+───────────────────────────────────────────────────────────────────────────────
+  RECOMMENDED SETTINGS
+───────────────────────────────────────────────────────────────────────────────
+  The author did not publish specific settings.
+  No author settings were published. Community baseline: steps Distilled
+  FLUX.2-klein-9B: 4 (HF card example; it is step-distilled to exactly this,
+  and more steps do not refine further). Base FLUX.2-klein-base-9B: 50 in the
+  HF card example and in ComfyUI's benchmark; community practice runs lower,
+  with ~20 for production and 8-12 for fast iteration reported in the CivitAI
+  consistency-LoRA article., cfg Distilled: guidance_scale 1.0 (HF card). It
+  is guidance-distilled and its transformer config sets
+  guidance_embeds=false, so raising CFG degrades rather than improves. Base:
+  guidance_scale 4.0 (HF card); the CivitAI consistency-LoRA article
+  recommends 3.5-4.5 for most edits., sampler Not stated by BFL or by
+  docs.comfy.org, and the official flux2 GitHub repo documents no sampler,
+  scheduler or shift values either. Community consensus is euler with a
+  resolution-dependent shifted-Euler / flow-match schedule (ComfyUI's Flux2
+  workflows use a scheduler that interpolates shift between base_shift and
+  max_shift by output resolution). Start from ComfyUI's downloadable Flux.2
+  klein workflow template rather than a blog. See unverified., resolution
+  1024x1024 in every BFL HF card example and in CivitAI's 9B training recipe
+  ("resolution": 1024). Non-square ratios work but no official list of
+  supported dimensions or a stated alignment multiple was found — see
+  unverified., lora_strength 1.0 is the diffusers/ComfyUI default and what
+  CivitAI applies to training previews (strength 1.0). Published per-LoRA
+  guidance varies and should be read from the LoRA, not assumed: the widely
+  used Klein 9B consistency LoRA's CivitAI article says start at 0.5 within a
+  0.2-1.0 range (0.2-0.4 loose, 0.5-0.7 balanced, 0.8-1.0 maximum structural
+  stability) and warns that too high a strength stops edits from applying;
+  SOLRICKS' Realistic Detail LoRA card says 0.6-0.8 subtle, 0.8-1.0 stronger,
+  start 0.8; the vafipas663 distillation-delta LoRA card warns it "produces
+  static noise at strength > 0.5. And sometimes, you have to go as low as
+  0.1.".
+
+───────────────────────────────────────────────────────────────────────────────
+  ADOPTION & POPULARITY
+───────────────────────────────────────────────────────────────────────────────
+  Downloads               41,046
+  Rating                  100% positive (2878/2881)
+  Comments                0
+  Published               209 days ago
+  Download rate           9.4/day since release
+
+  Adoption rank           #5 of 150 in this collection  (Top tier)
+  Momentum rank           #37 of 150 by download rate  (Trending)
+
+  Adoption tier is the percentile of total downloads within this collection;
+  momentum is the percentile of downloads-per-day since release. Momentum is
+  ranked rather than measured against a fixed rate, because this base model
+  is itself new and every LoRA looks fast on an absolute scale. Both numbers
+  are CivitAI's, read on the scrape date at the foot of this file.
+
+───────────────────────────────────────────────────────────────────────────────
+  CHAINING / STACKING
+───────────────────────────────────────────────────────────────────────────────
+  Stacks with others      not stated
+
+  The author gave no chaining guidance. The rules below are the
+  general FLUX.2 [klein] 9B ones, not anything specific to this LoRA.
+
+  General FLUX.2 [klein] 9B rules that apply here (see REFERENCE.txt):
+
+  • Chain LoRAs by stringing LoraLoaderModelOnly nodes in series, model
+    output into model input. There is no CLIP side to route — klein LoRAs
+    are model-only because the Qwen3 text encoder is not trained (CivitAI's
+    recipe fixes trainTextEncoder: false).
+  • Only stack LoRAs of the same parameter size. 9B stacks with 9B; a 4B
+    LoRA in a 9B stack is an architectural mismatch (3072 vs 4096 hidden,
+    5+20 vs 8+24 blocks), not a quality question.
+  • Mixing a "Flux.2 Klein 9B" LoRA and a "Flux.2 Klein 9B-base" LoRA in
+    one stack is fine — those strings describe the training checkpoint, not
+    incompatible tensor layouts. Set the sampler for the checkpoint
+    actually loaded.
+  • Run base-trained LoRAs on the distilled checkpoint when you want speed:
+    HF's klein LoRA guide reports the adapter loads there and typically
+    gives better results than on base, at 4 steps.
+  • A base-to-turbo / distillation-extraction LoRA is a whole-model delta,
+    not a content LoRA. Put it on the base checkpoint only, and treat its
+    strength as a steps/CFG dial rather than a style knob. Do not add it on
+    top of the already-distilled checkpoint.
+  • Read each LoRA's own published strength range instead of defaulting
+    everything to 1.0 — klein LoRAs in the wild ship very different ones
+    (0.5 start for the consistency LoRA, 0.8 for SOLRICKS' detail LoRA,
+    under 0.5 and sometimes 0.1 for the vafipas663 distillation delta).
+  • With a consistency or edit LoRA in the stack, excessive strength
+    suppresses the prompt: the consistency LoRA's article says that if
+    edits stop applying, lower it before changing anything else.
+  • Prefer runtime LoRA patching over baking a LoRA into the checkpoint:
+    ComfyUI's Model Merge / Model Save nodes have not detected Flux2 Klein
+    models (4B or 9B) since a March 2026 update — issue #13637, still open.
+  • The author never states which weight family this targets. Two different
+    questions, two different answers. Within 9B (distilled vs base — the
+    two CivitAI strings): not a real incompatibility. Both load through the
+    same Flux2KleinPipeline and have the same repo layout; for the 4B pair,
+    where both transformer configs are public, the distil
+
+───────────────────────────────────────────────────────────────────────────────
+  PROS
+───────────────────────────────────────────────────────────────────────────────
+  + Widely adopted — 41,046 downloads. [source: CivitAI stats]
+  + Positively received — 2,878 thumbs-up, 3 down. [source: CivitAI stats]
+  + Explicit trigger word(s) — ArsMovieStill, 80s Fantasy Movie Still — so
+    the effect can be turned on and off from the prompt. [source: model
+    metadata]
+  + Compact at 79 MB — cheap to stack with other LoRAs. [source: file size]
+
+───────────────────────────────────────────────────────────────────────────────
+  CONS & CAVEATS
+───────────────────────────────────────────────────────────────────────────────
+  - Author labels this experimental — results vary shot to shot. [source:
+    description]
+  - Author never states which generation mode it was trained for; test
+    against your own workflow before relying on it. [source: description]
+
+───────────────────────────────────────────────────────────────────────────────
+  COMPATIBILITY WARNING
+───────────────────────────────────────────────────────────────────────────────
+  Two different questions, two different answers. Within 9B (distilled vs
+  base — the two CivitAI strings): not a real incompatibility. Both load
+  through the same Flux2KleinPipeline and have the same repo layout; for the
+  4B pair, where both transformer configs are public, the distilled and base
+  configs are byte-for-byte identical in every shape field, and the 9B pair
+  is built the same way. HF's own klein LoRA guide states the workflow
+  directly: "Distilled models are step-compressed for fast inference; you
+  train against the base checkpoint and the adapter still loads on the
+  distilled model afterward", and "Applying the LoRA on the distilled model
+  typically gives better results than the base model, and it's faster" (that
+  post is written against 4B, but the family structure is the same). So the
+  two CivitAI strings record which checkpoint the author trained or previewed
+  against, not a hard compatibility wall. The real risk is a silent settings
+  mismatch: load a base-trained LoRA on the distilled checkpoint while
+  keeping the author's 20-50 step / CFG ~4 preview settings and the output
+  degrades badly (distilled wants 4 steps / guidance 1.0), or the reverse.
+  Nothing errors — it just looks wrong. Across sizes (9B vs 4B): a genuine
+  architectural mismatch, from the published transformer configs — different
+  hidden width (4096 vs 3072), different block counts (8+24 vs 5+20) and a
+  different text-conditioning dim (12288 vs 7680, following Qwen3-8B vs
+  Qwen3-4B). A 4B LoRA cannot patch a 9B model: expect missing keys and shape
+  errors, not degraded output. The exact ComfyUI error text is unverified. KV
+  variant: same 9B size and same 4-step distillation, but a distinct
+  checkpoint with its own pipeline; whether standard klein 9B LoRAs patch it
+  cleanly is not documented — see unverified.
+
+───────────────────────────────────────────────────────────────────────────────
+  AUTHOR'S DESCRIPTION
+───────────────────────────────────────────────────────────────────────────────
+  This LoRA is designed to capture the distinct look and feel of early 80s
+  fantasy movies.
+
+  The trigger words are ArsMovieStill, 80s Fantasy Movie Still
+  Z-IMAGE UPDATE: V2 dataset officially became my go-to test dataset when
+  training a model for the first time. Experimental!
+  Default ComfyUI LoRA Loader Reports Errors, but actually applys the LoRA.
+  Recommended usage settings:
+  - Your favourite spot.
+
+  - Listen to Castle Rat as background.
+
+  - Enjoy!
+
+  MJ7 UPDATE: Building each dataset is a joy, so with the release of
+  MidJourney 7, I naturally started this one again from scratch.
+  This version stays true to the original aesthetic but with improved
+  fidelity and an even stronger 80s vibe.
+  It’s trained on a fresh, carefully curated set of ~250 images, created by
+  referencing the best results from the previous versions.
+  V3 UPDATE: The training dataset has been further expanded to 1,000 curated
+  images, pushing it to the limits of Civit’s on-site training capabilities.
+  This update include broader selection of fantasy environments, mythical
+  creatures, and epic scenes, enhancing the fidelity to the iconic 80s
+  fantasy aesthetic. V2 UPDATE: Added to the dataset a huge amount of Fantasy
+  Characters, beasts and scenes.
+  Use v1 for stronger effects and v2/v3 for better versatility.
+
+───────────────────────────────────────────────────────────────────────────────
+  VERSION NOTES
+───────────────────────────────────────────────────────────────────────────────
+  V2 Dataset The effects are much stronger with the base than the Distilled
+  model. You can safely use 1.5 Str with the Distilled and still get a good
+  results.
+
+───────────────────────────────────────────────────────────────────────────────
+  EXAMPLE MEDIA
+───────────────────────────────────────────────────────────────────────────────
+  6 preview(s) in ./example_images/ — community results for
+  this LoRA. Videos keep a matching _poster.jpg still frame.
+
+  No prompts are recorded below: CivitAI's public API flags these posts as
+  having generation metadata but does not return it, so the prompts cannot be
+  scraped. Open the model page to read them.
+
+  01_image.jpeg  (image, 1024x1024)
+  02_image.jpeg  (image, 832x1216)
+  03_image.jpeg  (image, 1024x1024)
+  04_image.jpeg  (image, 1024x1024)
+  05_image.jpeg  (image, 1024x1024)
+  06_image.jpeg  (image, 1024x1024)
+
+───────────────────────────────────────────────────────────────────────────────
+  HOW TO USE (COMFYUI)
+───────────────────────────────────────────────────────────────────────────────
+  1. Diffusion model -> ComfyUI/models/diffusion_models/. Distilled:
+  flux-2-klein-9b-fp8.safetensors. Base: flux-2-klein-
+  base-9b-fp8.safetensors. (BF16 single-file weights
+  flux-2-klein-9b.safetensors / flux-2-klein-base-9b.safetensors from the
+  gated HF repos also work if you have ~29GB; community GGUF builds exist
+  from unsloth and leejet for lower-VRAM setups.)
+  2. Text encoder ->
+  ComfyUI/models/text_encoders/qwen_3_8b_fp8mixed.safetensors (from Comfy-
+  Org/vae-text-encorder-for-flux-klein-9b). This is the 9B family's encoder;
+  the 4B family uses qwen_3_4b.safetensors — do not mix them.
+  3. VAE -> ComfyUI/models/vae/flux2-vae.safetensors (shared across klein
+  sizes).
+  4. LoRA -> ComfyUI/models/loras/ as a plain .safetensors.
+  5. Load with Load Diffusion Model (UNETLoader) for the transformer,
+  CLIPLoader for the Qwen3 text encoder, VAELoader for the VAE.
+  docs.comfy.org ships prebuilt Flux.2 klein graphs under Workflow Templates
+  for both the distilled and base branches — start from those rather than
+  wiring by hand, since neither the docs nor BFL's repo state
+  sampler/scheduler/shift in prose; those values live in the template.
+  6. Insert LoraLoaderModelOnly between the diffusion-model loader and the
+  sampler. Model-only is correct for klein: the Qwen3 text encoder is not
+  trained by the standard klein trainers, so there is no text-encoder half to
+  patch.
+  7. Set steps and guidance to match the checkpoint you actually loaded, not
+  the one the LoRA was trained on: 4 steps / guidance 1.0 on distilled,
+  ~20-50 steps / guidance ~4.0 on base.
+  8. Note the 9B repos are gated on HuggingFace — you must accept the FLUX
+  Non-Commercial License on the repo page before any download, including
+  scripted ones.
+
+───────────────────────────────────────────────────────────────────────────────
+  Scraped by lorakit on 2026-09-05 01:47 UTC.
+  Stats and description are the author's; pros/cons are derived from the
+  evidence tagged beside each line.
+───────────────────────────────────────────────────────────────────────────────

80s-fantasy-movie/metadata.json

@@ -0,0 +1,138 @@
+{
+  "name": "80s Fantasy Movie",
+  "slug": "80s-fantasy-movie",
+  "base_model": "FLUX.2 [klein] 9B",
+  "media_type": "image",
+  "model_id": 789313,
+  "version_id": 2622285,
+  "version_name": "Klein9B",
+  "source": "civitai",
+  "url": "https://civitai.com/models/789313",
+  "repo_id": null,
+  "file_path": null,
+  "author": "ArsMachina",
+  "category": {
+    "key": "style",
+    "label": "Visual style"
+  },
+  "modalities": [],
+  "primary_modality": "unspecified",
+  "checkpoint_family": "unknown",
+  "trigger_words": [
+    "ArsMovieStill, 80s Fantasy Movie Still"
+  ],
+  "recommended_settings": {},
+  "chaining": {
+    "author": {
+      "stacks": "not stated",
+      "notes": "",
+      "avoid": [],
+      "requires": [],
+      "order": "",
+      "strength": "",
+      "stated": false
+    },
+    "general_rules": [
+      "Chain LoRAs by stringing LoraLoaderModelOnly nodes in series, model output into model input. There is no CLIP side to route — klein LoRAs are model-only because the Qwen3 text encoder is not trained (CivitAI's recipe fixes trainTextEncoder: false).",
+      "Only stack LoRAs of the same parameter size. 9B stacks with 9B; a 4B LoRA in a 9B stack is an architectural mismatch (3072 vs 4096 hidden, 5+20 vs 8+24 blocks), not a quality question.",
+      "Mixing a \"Flux.2 Klein 9B\" LoRA and a \"Flux.2 Klein 9B-base\" LoRA in one stack is fine — those strings describe the training checkpoint, not incompatible tensor layouts. Set the sampler for the checkpoint actually loaded.",
+      "Run base-trained LoRAs on the distilled checkpoint when you want speed: HF's klein LoRA guide reports the adapter loads there and typically gives better results than on base, at 4 steps.",
+      "A base-to-turbo / distillation-extraction LoRA is a whole-model delta, not a content LoRA. Put it on the base checkpoint only, and treat its strength as a steps/CFG dial rather than a style knob. Do not add it on top of the already-distilled checkpoint.",
+      "Read each LoRA's own published strength range instead of defaulting everything to 1.0 — klein LoRAs in the wild ship very different ones (0.5 start for the consistency LoRA, 0.8 for SOLRICKS' detail LoRA, under 0.5 and sometimes 0.1 for the vafipas663 distillation delta).",
+      "With a consistency or edit LoRA in the stack, excessive strength suppresses the prompt: the consistency LoRA's article says that if edits stop applying, lower it before changing anything else.",
+      "Prefer runtime LoRA patching over baking a LoRA into the checkpoint: ComfyUI's Model Merge / Model Save nodes have not detected Flux2 Klein models (4B or 9B) since a March 2026 update — issue #13637, still open.",
+      "The author never states which weight family this targets. Two different questions, two different answers. Within 9B (distilled vs base — the two CivitAI strings): not a real incompatibility. Both load through the same Flux2KleinPipeline and have the same repo layout; for the 4B pair, where both transformer configs are public, the distil"
+    ]
+  },
+  "pros": [
+    "Widely adopted — 41,046 downloads. [source: CivitAI stats]",
+    "Positively received — 2,878 thumbs-up, 3 down. [source: CivitAI stats]",
+    "Explicit trigger word(s) — ArsMovieStill, 80s Fantasy Movie Still — so the effect can be turned on and off from the prompt. [source: model metadata]",
+    "Compact at 79 MB — cheap to stack with other LoRAs. [source: file size]"
+  ],
+  "cons": [
+    "Author labels this experimental — results vary shot to shot. [source: description]",
+    "Author never states which generation mode it was trained for; test against your own workflow before relying on it. [source: description]"
+  ],
+  "file": {
+    "name": "80sFantasyKlein9b_000002000H1000.safetensors",
+    "size_bytes": 82866712,
+    "size_mb": 79.03,
+    "sha256": "4db57a092e5ac1ec1dade5a6fd6c645f173a6e97c9cac425a8ddbb7689e34a0a"
+  },
+  "stats": {
+    "downloads": 41046,
+    "thumbs_up": 2878,
+    "thumbs_down": 3
+  },
+  "popularity": {
+    "downloads": 41046,
+    "thumbs_up": 2878,
+    "thumbs_down": 3,
+    "comments": 0,
+    "approval": "100% positive (2878/2881)",
+    "age_days": 208.6,
+    "downloads_per_day": 9.36,
+    "rank_by_downloads": 5,
+    "rank_by_momentum": 37,
+    "of_total": 150,
+    "adoption_tier": "Top tier",
+    "momentum": "Trending"
+  },
+  "selection_score": 75982.0,
+  "selection_reasons": [
+    "41,046 downloads",
+    "2,878 thumbs-up",
+    "documented triggers + usage notes",
+    "top-ranked for its category"
+  ],
+  "example_media": [
+    {
+      "file": "01_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "02_image.jpeg",
+      "type": "image",
+      "resolution": "832x1216",
+      "prompt": ""
+    },
+    {
+      "file": "03_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "04_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "05_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "06_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    }
+  ],
+  "tags": [
+    "80s",
+    "fantasy",
+    "movie",
+    "still",
+    "arsmjstyle",
+    "midjourney v6",
+    "style"
+  ],
+  "scraped_at": "2026-09-05T01:47:42+00:00",
+  "path": "flux2-klein-9b/80s-fantasy-movie"
+}

art-nouveau/ArtNoveauKlein9b.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87cce20088c6734cc4ee4cbea3631b53af4ef3d3d8a00810df47f04853b9f90f
+size 82866712

art-nouveau/info.txt

@@ -0,0 +1,253 @@
+═══════════════════════════════════════════════════════════════════════════════
+  ART NOUVEAU
+═══════════════════════════════════════════════════════════════════════════════
+
+───────────────────────────────────────────────────────────────────────────────
+  IDENTITY
+───────────────────────────────────────────────────────────────────────────────
+  Base model        FLUX.2 [klein] 9B
+  Generation mode   not declared by the author
+  Checkpoint        not determined — verify before use
+  Category          Visual style
+  Author            ArsMachina
+  Source            https://civitai.com/models/562604
+  Provenance        CivitAI · model 562604 · version 2760648
+
+───────────────────────────────────────────────────────────────────────────────
+  WEIGHTS
+───────────────────────────────────────────────────────────────────────────────
+  File              ArtNoveauKlein9b.safetensors
+  Version           Klein9B
+  Size              79.0 MB
+  Format            SafeTensor / unknown precision
+  SHA256            87cce20088c6734cc4ee4cbea3631b53af4ef3d3d8a00810df47f04853b9f90f
+  Published         2026-04-19
+
+───────────────────────────────────────────────────────────────────────────────
+  TRIGGER WORDS
+───────────────────────────────────────────────────────────────────────────────
+  › "ArsMJStyle, art nouveau,"
+
+───────────────────────────────────────────────────────────────────────────────
+  RECOMMENDED SETTINGS
+───────────────────────────────────────────────────────────────────────────────
+  The author did not publish specific settings.
+  No author settings were published. Community baseline: steps Distilled
+  FLUX.2-klein-9B: 4 (HF card example; it is step-distilled to exactly this,
+  and more steps do not refine further). Base FLUX.2-klein-base-9B: 50 in the
+  HF card example and in ComfyUI's benchmark; community practice runs lower,
+  with ~20 for production and 8-12 for fast iteration reported in the CivitAI
+  consistency-LoRA article., cfg Distilled: guidance_scale 1.0 (HF card). It
+  is guidance-distilled and its transformer config sets
+  guidance_embeds=false, so raising CFG degrades rather than improves. Base:
+  guidance_scale 4.0 (HF card); the CivitAI consistency-LoRA article
+  recommends 3.5-4.5 for most edits., sampler Not stated by BFL or by
+  docs.comfy.org, and the official flux2 GitHub repo documents no sampler,
+  scheduler or shift values either. Community consensus is euler with a
+  resolution-dependent shifted-Euler / flow-match schedule (ComfyUI's Flux2
+  workflows use a scheduler that interpolates shift between base_shift and
+  max_shift by output resolution). Start from ComfyUI's downloadable Flux.2
+  klein workflow template rather than a blog. See unverified., resolution
+  1024x1024 in every BFL HF card example and in CivitAI's 9B training recipe
+  ("resolution": 1024). Non-square ratios work but no official list of
+  supported dimensions or a stated alignment multiple was found — see
+  unverified., lora_strength 1.0 is the diffusers/ComfyUI default and what
+  CivitAI applies to training previews (strength 1.0). Published per-LoRA
+  guidance varies and should be read from the LoRA, not assumed: the widely
+  used Klein 9B consistency LoRA's CivitAI article says start at 0.5 within a
+  0.2-1.0 range (0.2-0.4 loose, 0.5-0.7 balanced, 0.8-1.0 maximum structural
+  stability) and warns that too high a strength stops edits from applying;
+  SOLRICKS' Realistic Detail LoRA card says 0.6-0.8 subtle, 0.8-1.0 stronger,
+  start 0.8; the vafipas663 distillation-delta LoRA card warns it "produces
+  static noise at strength > 0.5. And sometimes, you have to go as low as
+  0.1.".
+
+───────────────────────────────────────────────────────────────────────────────
+  ADOPTION & POPULARITY
+───────────────────────────────────────────────────────────────────────────────
+  Downloads               14,624
+  Rating                  100% positive (1172/1174)
+  Comments                0
+  Published               139 days ago
+  Download rate           3.3/day since release
+
+  Adoption rank           #26 of 150 in this collection  (High)
+  Momentum rank           #100 of 150 by download rate  (Long tail)
+
+  Adoption tier is the percentile of total downloads within this collection;
+  momentum is the percentile of downloads-per-day since release. Momentum is
+  ranked rather than measured against a fixed rate, because this base model
+  is itself new and every LoRA looks fast on an absolute scale. Both numbers
+  are CivitAI's, read on the scrape date at the foot of this file.
+
+───────────────────────────────────────────────────────────────────────────────
+  CHAINING / STACKING
+───────────────────────────────────────────────────────────────────────────────
+  Stacks with others      not stated
+
+  The author gave no chaining guidance. The rules below are the
+  general FLUX.2 [klein] 9B ones, not anything specific to this LoRA.
+
+  General FLUX.2 [klein] 9B rules that apply here (see REFERENCE.txt):
+
+  • Chain LoRAs by stringing LoraLoaderModelOnly nodes in series, model
+    output into model input. There is no CLIP side to route — klein LoRAs
+    are model-only because the Qwen3 text encoder is not trained (CivitAI's
+    recipe fixes trainTextEncoder: false).
+  • Only stack LoRAs of the same parameter size. 9B stacks with 9B; a 4B
+    LoRA in a 9B stack is an architectural mismatch (3072 vs 4096 hidden,
+    5+20 vs 8+24 blocks), not a quality question.
+  • Mixing a "Flux.2 Klein 9B" LoRA and a "Flux.2 Klein 9B-base" LoRA in
+    one stack is fine — those strings describe the training checkpoint, not
+    incompatible tensor layouts. Set the sampler for the checkpoint
+    actually loaded.
+  • Run base-trained LoRAs on the distilled checkpoint when you want speed:
+    HF's klein LoRA guide reports the adapter loads there and typically
+    gives better results than on base, at 4 steps.
+  • A base-to-turbo / distillation-extraction LoRA is a whole-model delta,
+    not a content LoRA. Put it on the base checkpoint only, and treat its
+    strength as a steps/CFG dial rather than a style knob. Do not add it on
+    top of the already-distilled checkpoint.
+  • Read each LoRA's own published strength range instead of defaulting
+    everything to 1.0 — klein LoRAs in the wild ship very different ones
+    (0.5 start for the consistency LoRA, 0.8 for SOLRICKS' detail LoRA,
+    under 0.5 and sometimes 0.1 for the vafipas663 distillation delta).
+  • With a consistency or edit LoRA in the stack, excessive strength
+    suppresses the prompt: the consistency LoRA's article says that if
+    edits stop applying, lower it before changing anything else.
+  • Prefer runtime LoRA patching over baking a LoRA into the checkpoint:
+    ComfyUI's Model Merge / Model Save nodes have not detected Flux2 Klein
+    models (4B or 9B) since a March 2026 update — issue #13637, still open.
+  • The author never states which weight family this targets. Two different
+    questions, two different answers. Within 9B (distilled vs base — the
+    two CivitAI strings): not a real incompatibility. Both load through the
+    same Flux2KleinPipeline and have the same repo layout; for the 4B pair,
+    where both transformer configs are public, the distil
+
+───────────────────────────────────────────────────────────────────────────────
+  PROS
+───────────────────────────────────────────────────────────────────────────────
+  + Widely adopted — 14,624 downloads. [source: CivitAI stats]
+  + Positively received — 1,172 thumbs-up, 2 down. [source: CivitAI stats]
+  + Explicit trigger word(s) — ArsMJStyle, art nouveau, — so the effect can
+    be turned on and off from the prompt. [source: model metadata]
+  + Compact at 79 MB — cheap to stack with other LoRAs. [source: file size]
+
+───────────────────────────────────────────────────────────────────────────────
+  CONS & CAVEATS
+───────────────────────────────────────────────────────────────────────────────
+  - Author never states which generation mode it was trained for; test
+    against your own workflow before relying on it. [source: description]
+
+───────────────────────────────────────────────────────────────────────────────
+  COMPATIBILITY WARNING
+───────────────────────────────────────────────────────────────────────────────
+  Two different questions, two different answers. Within 9B (distilled vs
+  base — the two CivitAI strings): not a real incompatibility. Both load
+  through the same Flux2KleinPipeline and have the same repo layout; for the
+  4B pair, where both transformer configs are public, the distilled and base
+  configs are byte-for-byte identical in every shape field, and the 9B pair
+  is built the same way. HF's own klein LoRA guide states the workflow
+  directly: "Distilled models are step-compressed for fast inference; you
+  train against the base checkpoint and the adapter still loads on the
+  distilled model afterward", and "Applying the LoRA on the distilled model
+  typically gives better results than the base model, and it's faster" (that
+  post is written against 4B, but the family structure is the same). So the
+  two CivitAI strings record which checkpoint the author trained or previewed
+  against, not a hard compatibility wall. The real risk is a silent settings
+  mismatch: load a base-trained LoRA on the distilled checkpoint while
+  keeping the author's 20-50 step / CFG ~4 preview settings and the output
+  degrades badly (distilled wants 4 steps / guidance 1.0), or the reverse.
+  Nothing errors — it just looks wrong. Across sizes (9B vs 4B): a genuine
+  architectural mismatch, from the published transformer configs — different
+  hidden width (4096 vs 3072), different block counts (8+24 vs 5+20) and a
+  different text-conditioning dim (12288 vs 7680, following Qwen3-8B vs
+  Qwen3-4B). A 4B LoRA cannot patch a 9B model: expect missing keys and shape
+  errors, not degraded output. The exact ComfyUI error text is unverified. KV
+  variant: same 9B size and same 4-step distillation, but a distinct
+  checkpoint with its own pipeline; whether standard klein 9B LoRAs patch it
+  cleanly is not documented — see unverified.
+
+───────────────────────────────────────────────────────────────────────────────
+  AUTHOR'S DESCRIPTION
+───────────────────────────────────────────────────────────────────────────────
+  The Lora is trained on 140 MidJourney Images that use my "Personalised
+  Style" + Art Nouveau The dataset is diverse and tries to capture the true
+  spirit of Art Nouveau seen through my lens, not just portraits in Alphonse
+  Mucha style.
+  Works without the trigger words ArsMJStyle, Art Nouveau, but gives stronger
+  results with them.
+
+  MJ7 VERSION: The dataset was rebuilt entirely from scratch using MidJourney
+  V7, once again mixed with my personalized style.
+  While the overall composition is similar to V2, this version expands the
+  range by including everyday objects, architectural details, and jewelry.
+  Expect improved structure and more natural colors.
+  V2 Changes :
+  I've completely rebuilt the dataset, to better reflect the evolution of my
+  MidJourney style and LoRA structuring/labeling I use.
+  This new version (v2) is based on a curated dataset of 250 images :
+  - 100 Portraits
+
+  - 150 Landscapes, Buildings, Animals, and Abstract Paintings
+
+  The colors and composition in v2 offer a distinct experience compared to
+  v1, so I encourage you to try both versions and see which one you like the
+  most.
+  Recommended settings remain the same as v1.
+
+───────────────────────────────────────────────────────────────────────────────
+  EXAMPLE MEDIA
+───────────────────────────────────────────────────────────────────────────────
+  6 preview(s) in ./example_images/ — community results for
+  this LoRA. Videos keep a matching _poster.jpg still frame.
+
+  No prompts are recorded below: CivitAI's public API flags these posts as
+  having generation metadata but does not return it, so the prompts cannot be
+  scraped. Open the model page to read them.
+
+  01_image.jpeg  (image, 1440x2560)
+  02_image.jpeg  (image, 1024x1024)
+  03_image.jpeg  (image, 1024x1024)
+  04_image.jpeg  (image, 1024x1024)
+  05_image.jpeg  (image, 1440x2560)
+  06_image.jpeg  (image, 1440x2560)
+
+───────────────────────────────────────────────────────────────────────────────
+  HOW TO USE (COMFYUI)
+───────────────────────────────────────────────────────────────────────────────
+  1. Diffusion model -> ComfyUI/models/diffusion_models/. Distilled:
+  flux-2-klein-9b-fp8.safetensors. Base: flux-2-klein-
+  base-9b-fp8.safetensors. (BF16 single-file weights
+  flux-2-klein-9b.safetensors / flux-2-klein-base-9b.safetensors from the
+  gated HF repos also work if you have ~29GB; community GGUF builds exist
+  from unsloth and leejet for lower-VRAM setups.)
+  2. Text encoder ->
+  ComfyUI/models/text_encoders/qwen_3_8b_fp8mixed.safetensors (from Comfy-
+  Org/vae-text-encorder-for-flux-klein-9b). This is the 9B family's encoder;
+  the 4B family uses qwen_3_4b.safetensors — do not mix them.
+  3. VAE -> ComfyUI/models/vae/flux2-vae.safetensors (shared across klein
+  sizes).
+  4. LoRA -> ComfyUI/models/loras/ as a plain .safetensors.
+  5. Load with Load Diffusion Model (UNETLoader) for the transformer,
+  CLIPLoader for the Qwen3 text encoder, VAELoader for the VAE.
+  docs.comfy.org ships prebuilt Flux.2 klein graphs under Workflow Templates
+  for both the distilled and base branches — start from those rather than
+  wiring by hand, since neither the docs nor BFL's repo state
+  sampler/scheduler/shift in prose; those values live in the template.
+  6. Insert LoraLoaderModelOnly between the diffusion-model loader and the
+  sampler. Model-only is correct for klein: the Qwen3 text encoder is not
+  trained by the standard klein trainers, so there is no text-encoder half to
+  patch.
+  7. Set steps and guidance to match the checkpoint you actually loaded, not
+  the one the LoRA was trained on: 4 steps / guidance 1.0 on distilled,
+  ~20-50 steps / guidance ~4.0 on base.
+  8. Note the 9B repos are gated on HuggingFace — you must accept the FLUX
+  Non-Commercial License on the repo page before any download, including
+  scripted ones.
+
+───────────────────────────────────────────────────────────────────────────────
+  Scraped by lorakit on 2026-09-05 01:48 UTC.
+  Stats and description are the author's; pros/cons are derived from the
+  evidence tagged beside each line.
+───────────────────────────────────────────────────────────────────────────────

art-nouveau/metadata.json

@@ -0,0 +1,135 @@
+{
+  "name": "Art Nouveau",
+  "slug": "art-nouveau",
+  "base_model": "FLUX.2 [klein] 9B",
+  "media_type": "image",
+  "model_id": 562604,
+  "version_id": 2760648,
+  "version_name": "Klein9B",
+  "source": "civitai",
+  "url": "https://civitai.com/models/562604",
+  "repo_id": null,
+  "file_path": null,
+  "author": "ArsMachina",
+  "category": {
+    "key": "style",
+    "label": "Visual style"
+  },
+  "modalities": [],
+  "primary_modality": "unspecified",
+  "checkpoint_family": "unknown",
+  "trigger_words": [
+    "ArsMJStyle, art nouveau,"
+  ],
+  "recommended_settings": {},
+  "chaining": {
+    "author": {
+      "stacks": "not stated",
+      "notes": "",
+      "avoid": [],
+      "requires": [],
+      "order": "",
+      "strength": "",
+      "stated": false
+    },
+    "general_rules": [
+      "Chain LoRAs by stringing LoraLoaderModelOnly nodes in series, model output into model input. There is no CLIP side to route — klein LoRAs are model-only because the Qwen3 text encoder is not trained (CivitAI's recipe fixes trainTextEncoder: false).",
+      "Only stack LoRAs of the same parameter size. 9B stacks with 9B; a 4B LoRA in a 9B stack is an architectural mismatch (3072 vs 4096 hidden, 5+20 vs 8+24 blocks), not a quality question.",
+      "Mixing a \"Flux.2 Klein 9B\" LoRA and a \"Flux.2 Klein 9B-base\" LoRA in one stack is fine — those strings describe the training checkpoint, not incompatible tensor layouts. Set the sampler for the checkpoint actually loaded.",
+      "Run base-trained LoRAs on the distilled checkpoint when you want speed: HF's klein LoRA guide reports the adapter loads there and typically gives better results than on base, at 4 steps.",
+      "A base-to-turbo / distillation-extraction LoRA is a whole-model delta, not a content LoRA. Put it on the base checkpoint only, and treat its strength as a steps/CFG dial rather than a style knob. Do not add it on top of the already-distilled checkpoint.",
+      "Read each LoRA's own published strength range instead of defaulting everything to 1.0 — klein LoRAs in the wild ship very different ones (0.5 start for the consistency LoRA, 0.8 for SOLRICKS' detail LoRA, under 0.5 and sometimes 0.1 for the vafipas663 distillation delta).",
+      "With a consistency or edit LoRA in the stack, excessive strength suppresses the prompt: the consistency LoRA's article says that if edits stop applying, lower it before changing anything else.",
+      "Prefer runtime LoRA patching over baking a LoRA into the checkpoint: ComfyUI's Model Merge / Model Save nodes have not detected Flux2 Klein models (4B or 9B) since a March 2026 update — issue #13637, still open.",
+      "The author never states which weight family this targets. Two different questions, two different answers. Within 9B (distilled vs base — the two CivitAI strings): not a real incompatibility. Both load through the same Flux2KleinPipeline and have the same repo layout; for the 4B pair, where both transformer configs are public, the distil"
+    ]
+  },
+  "pros": [
+    "Widely adopted — 14,624 downloads. [source: CivitAI stats]",
+    "Positively received — 1,172 thumbs-up, 2 down. [source: CivitAI stats]",
+    "Explicit trigger word(s) — ArsMJStyle, art nouveau, — so the effect can be turned on and off from the prompt. [source: model metadata]",
+    "Compact at 79 MB — cheap to stack with other LoRAs. [source: file size]"
+  ],
+  "cons": [
+    "Author never states which generation mode it was trained for; test against your own workflow before relying on it. [source: description]"
+  ],
+  "file": {
+    "name": "ArtNoveauKlein9b.safetensors",
+    "size_bytes": 82866712,
+    "size_mb": 79.03,
+    "sha256": "87cce20088c6734cc4ee4cbea3631b53af4ef3d3d8a00810df47f04853b9f90f"
+  },
+  "stats": {
+    "downloads": 14624,
+    "thumbs_up": 1172,
+    "thumbs_down": 2
+  },
+  "popularity": {
+    "downloads": 14624,
+    "thumbs_up": 1172,
+    "thumbs_down": 2,
+    "comments": 0,
+    "approval": "100% positive (1172/1174)",
+    "age_days": 138.7,
+    "downloads_per_day": 3.27,
+    "rank_by_downloads": 26,
+    "rank_by_momentum": 100,
+    "of_total": 150,
+    "adoption_tier": "High",
+    "momentum": "Long tail"
+  },
+  "selection_score": 29088.0,
+  "selection_reasons": [
+    "14,624 downloads",
+    "1,172 thumbs-up",
+    "documented triggers + usage notes",
+    "top-ranked for its category"
+  ],
+  "example_media": [
+    {
+      "file": "01_image.jpeg",
+      "type": "image",
+      "resolution": "1440x2560",
+      "prompt": ""
+    },
+    {
+      "file": "02_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "03_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "04_image.jpeg",
+      "type": "image",
+      "resolution": "1024x1024",
+      "prompt": ""
+    },
+    {
+      "file": "05_image.jpeg",
+      "type": "image",
+      "resolution": "1440x2560",
+      "prompt": ""
+    },
+    {
+      "file": "06_image.jpeg",
+      "type": "image",
+      "resolution": "1440x2560",
+      "prompt": ""
+    }
+  ],
+  "tags": [
+    "midjourney",
+    "style",
+    "arsmjstyle",
+    "midjourney v6",
+    "art nouveau"
+  ],
+  "scraped_at": "2026-09-05T01:48:17+00:00",
+  "path": "flux2-klein-9b/art-nouveau"
+}