initial commit

.gitattributes

@@ -0,0 +1,55 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.lz4 filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz 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
+# Audio files - uncompressed
+*.pcm filter=lfs diff=lfs merge=lfs -text
+*.sam filter=lfs diff=lfs merge=lfs -text
+*.raw filter=lfs diff=lfs merge=lfs -text
+# Audio files - compressed
+*.aac filter=lfs diff=lfs merge=lfs -text
+*.flac filter=lfs diff=lfs merge=lfs -text
+*.mp3 filter=lfs diff=lfs merge=lfs -text
+*.ogg filter=lfs diff=lfs merge=lfs -text
+*.wav filter=lfs diff=lfs merge=lfs -text
+# Image files - uncompressed
+*.bmp filter=lfs diff=lfs merge=lfs -text
+*.gif filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.tiff filter=lfs diff=lfs merge=lfs -text
+# Image files - compressed
+*.jpg filter=lfs diff=lfs merge=lfs -text
+*.jpeg filter=lfs diff=lfs merge=lfs -text
+*.webp filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,132 @@
+---
+tags:
+- text-to-image
+- lora
+- diffusers
+- template:diffusion-lora
+widget:
+- output:
+    url: images/Image01_00646_.png
+  text: "Astrid,\uFEFF High-fashion editorial studio portrait, ultra-realistic professional photography. Fully human, natural appearance with true-to-life skin behavior and optical sharpness.\uFEFF Aspect ratio: 5:7\uFEFF Framing: Slightly below the bust, half-body portrait.\uFEFF Subject & Expression:\uFEFF A young woman photographed from the bust up, centered in the frame, looking directly into the camera with a confident, magnetic presence.\uFEFF Her facial expression is intense and subtly seductive. Lips slightly parted and gently projected forward, conveying sensual introspection. Eyes maintain direct contact with the camera, eyelids softly lowered to create a sense of languid mystery. Eyebrows are relaxed and natural, avoiding graphic stiffness.\uFEFF Her head is tilted slightly back and to the side, enhancing the dramatic, editorial pose.\uFEFF Pose & Body Language:\uFEFF The posture is slightly angled, with shoulders subtly turned forward.\uFEFF A gentle neck extension creates a strong visual line between face, neck, and shoulders, adding elegance, tension, and sculptural definition.\uFEFF Makeup (Editorial, Contemporary):\uFEFF • Skin: Natural skin texture preserved with visible micro-pores and realistic surface variation. Lightweight luminous foundation, subtle contouring, satin finish.\uFEFF • Eyes: Deep black smokey eye, precise cat-eye liner, elongated lashes. Lighting does not obscure lash definition.\uFEFF • Brows: Natural, softly defined, non-graphic.\uFEFF • Lips: Brown-nude creamy lipstick with a soft reflective sheen.\uFEFF Hair:\uFEFF Wet-look hairstyle with a clean center part. Hair appears slightly disheveled, with loose strands falling across the face. Intentional, raw, sensual, modern aesthetic.\uFEFF Wardrobe:\uFEFF Premium black leather overbust strapless corset with a deep open U-shaped neckline. Vertical boning and angular side-pointed panel construction. Clearly defined leather texture and structure.\uFEFF Accessories:\uFEFF Two chokers:\uFEFF – One thin to medium black leather choker without pendant\uFEFF – One slightly lower medium-thickness silver chain choker without pendant\uFEFF No earrings. Visible silver ear cuff and multiple small silver piercings (studs and hoops).\uFEFF Lighting & Environment:\uFEFF Dark low-key studio environment.\uFEFF A single circular spotlight with a red gel is projected onto the wall behind the model, forming a clean, well-defined red halo.\uFEFF The model stands between the light source and the wall, casting a realistic shadow inside the red circle.\uFEFF All surrounding areas remain unlit and dark.\uFEFF Lighting is organic and enveloping, with rounded volumes and smooth shadow transitions. Strong chiaroscuro contrast.\uFEFF Focus & Detail:\uFEFF Absolute critical focus on the face.\uFEFF High micro-contrast and optical clarity.\uFEFF Visible textures: leather grain, metallic jewelry, natural skin detail.\uFEFF Camera & Optics:\uFEFF Sony A7R IV\uFEFF 85mm lens\uFEFF ISO 100\uFEFF Aperture f/2.8\uFEFF Shutter speed 1/125\uFEFF Shallow depth of field with natural falloff.\uFEFF Composition & Color Palette:\uFEFF Centered composition.\uFEFF High contrast (light vs shadow).\uFEFF Color palette: deep blacks, vibrant red spotlight, silver metallic accents."
+- output:
+    url: images/Image01_00647_.png
+  text: "Astrid, A highly realistic, cinematic medium close-up portrait of a beautiful young woman with all original facial features preserved, wearing soft smokey eye makeup and nude glossy lips. She smiles gently with eyes closed, resting her chin gracefully on her crossed hands atop a massive luxurious bouquet of 101 white roses in full bloom. She wears a delicate off-shoulder white satin dress with thin straps, a sparkling diamond necklace, matching drop earrings, and a fine diamond bracelet on her left wrist. Her manicured hands with long almond-shaped nude-pink nails rest elegantly on the roses.\uFEFF    The setting is a modern minimalist luxury apartment bathed in natural golden-hour sunlight streaming in from the right, creating warm highlights on her face and the bouquet. The softly blurred background (bokeh) reveals a refined white interior — a low console table, modern lamp, wall-mounted TV, and a white sofa in the distance.\uFEFF    Captured with a Leica camera, 85mm lens, f/1.8 aperture. Hyper-realistic detail, soft depth of field, cinematic lighting, warm color grading, luxurious and romantic atmosphere."
+- output:
+    url: images/Image01_00650_.png
+  text: >-
+    A black-and-white high-fashion editorial portrait of a woman wearing a
+    loose, flowing white blouse. The fabric drapes softly over her frame,
+    catching light and shadow with sculptural elegance. She poses with one arm
+    raised, lifting and holding her hair while partially obscuring part of her
+    face, creating a dramatic and abstract composition that feels intentional
+    and expressive.  Her gaze is intense and unwavering, directed straight into
+    the camera, conveying a strong, enigmatic presence. The emotion feels raw
+    and controlled—quiet power rather than overt performance. Her expression,
+    combined with the partial obstruction of her face, adds mystery and
+    editorial tension.  The background is a minimalist plain wall, free of
+    distractions. Harsh natural daylight streams in from one side, casting
+    sharply defined shadows across the wall and her face. The lighting is
+    high-contrast and unforgiving, emphasizing sharp cheekbones, facial
+    contours, and the natural texture of her skin, hair, and fabric. Shadows
+    fall decisively, creating graphic lines and depth typical of classic fashion
+    editorials.  The camera is positioned at eye level, framed tightly as a
+    close-up to prioritize facial expression, texture, and shadow interplay. The
+    image is captured with a full-frame DSLR using an 85mm lens at a wide
+    aperture, producing crisp detail, subtle tonal depth, and fine grain.  The
+    overall style is bold, minimalist, and editorial—pure monochrome fashion
+    photography with a strong emphasis on form, light, and emotion.
+- output:
+    url: images/Image01_00651_.png
+  text: '-'
+- output:
+    url: images/Image01_00652_.png
+  text: "A high-resolution, high-end editorial portrait of a young slim woman. The model is bathed in golden sunlight, captured from a low-angle shot to emphasize confidence and poise. Shot on a 50mm lens with a wide aperture (f/1.8), creating a beautifully blurred background of soft Mediterranean-style buildings that evoke a late-afternoon urban warmth.\uFEFF \_ She stands slightly tilted forward, chin subtly raised, eyes half-closed as if savoring the sun, lips relaxed in quiet self-assurance. She wears a tailored mustard yellow blazer layered over a lace bodice top with fine framing details, paired with chunky gold hoop earrings and a delicate chain necklace — modern yet timeless. Her short tousled hair catches the light naturally, adding movement and authenticity.\uFEFF \_ The image features warm golden-hour color grading with sunlit highlights and a gentle film-grain texture, achieving an editorial, cinematic finish that radiates effortless sophistication — sunlit, confident, and boldly feminine."
+base_model: Tongyi-MAI/Z-Image-Turbo
+instance_prompt: Astrid
+license: apache-2.0
+---
+# Àstrid Bergès-Frisbey
+
+<Gallery />
+
+## Model description 
+
+🧩 **LoRA Name:** **Astrid_ZimageTurbo**
+
+&gt; Base Model: Tongyi-MAI &#x2F; Z-Image-Turbo
+&gt; LoRA Type: Character &#x2F; Actress
+&gt; Trigger Word(s): **Astrid**
+&gt; Recommended Weight: **0.6 – 1.0** (ideal range ≈ **0.8–0.9**)
+
+---
+
+✨ **Overview**
+
+&gt; This LoRA is trained to capture the distinctive identity, facial structure, and ethereal presence of **Astrid Bergès-Frisbey**, optimized specifically for **Z-Image-Turbo**. The model emphasizes likeness fidelity, subtle sensuality, and editorial realism, with a strong bias toward cinematic, European, and high-fashion aesthetics.
+
+It excels at portraying Astrid’s delicate bone structure, intense gaze, and understated elegance across studio editorials, natural-light urban scenes, luxury interiors, and intimate portrait compositions.
+
+---
+
+🎨 **What This LoRA Is Good At**
+
+&gt; High likeness accuracy with preserved facial geometry
+&gt; Moody, cinematic, and editorial portraiture
+&gt; Natural skin texture with realistic micro-detail
+&gt; European fashion and lifestyle aesthetics
+&gt; Strong performance in low-key lighting, chiaroscuro, and soft daylight
+&gt; Portrait, half-body, and mirror-selfie compositions
+
+---
+
+⚙️ **Usage Notes**
+
+&gt; Use the trigger word **“Astrid”** once in the prompt
+&gt; Avoid stacking other character LoRAs to preserve identity fidelity
+&gt; Performs best with realistic lenses (50–85mm) and fashion&#x2F;editorial lighting
+&gt; Guidance scale **≤ 1** recommended for native Z-Image-Turbo behavior
+&gt; Responds well to detailed wardrobe, makeup, and lighting descriptions
+
+---
+
+🧪 **Example Prompt**
+Astrid, cinematic editorial portrait, soft smokey eyes,
+natural skin texture, low-key studio lighting,
+European fashion aesthetic, ultra-realistic
+
+---
+
+📐 **Training Details**
+
+&gt; Dataset: Captioned image dataset
+&gt; Training Resolutions: 512 &#x2F; 768
+&gt; Total Steps: 3,500
+&gt; Batch Size: 1
+&gt; Content vs Style: Balanced
+&gt; Noise Scheduler: FlowMatch
+&gt; Optimizer: AdamW 8-bit
+&gt; Learning Rate: 1e-4
+&gt; Precision: BF16
+&gt; Text Encoder Training: Disabled
+&gt; UNet Training: Enabled
+&gt; Quantization: QFloat8 (model + text encoder)
+&gt; Architecture: ZImage Turbo LoRA (LoKR full-rank enabled)
+
+---
+
+⚠️ **Limitations &#x2F; Notes**
+
+&gt; Not designed for anime or heavily stylized&#x2F;cartoon outputs
+&gt; Extreme facial exaggeration or ultra-wide lenses may reduce likeness
+&gt; Best results achieved when paired with realistic photography prompts
+
+
+## Trigger words
+
+You should use `Astrid` to trigger the image generation.
+
+
+## Download model
+
+
+[Download](/JunkieMonkey69/AstridBergesFrisbey_ZimageTurbo/tree/main) them in the Files & versions tab.