focusslider h3
Bandingkan
8281°
Autoplay
Jalankan secara tempatan
MMH3-FocusSlider-V1.safetensors
1 MB · SafeTensor · bf16
Muat turun disediakan oleh Civitai. Sesetengah pencipta memerlukan log masuk Civitai dahulu.
Tentang model ini
MINIMAX H3 · LoRA
by alcaitiff - [MMH3] FocusSlider - Negative values → very shallow depth of field. Positive values → deep focus / everything sharp
Nota pencipta
A tiny LoRA with a very specific job: controlling how much of the scene is in focus.
It’s only rank 1, around 650 KB, and was trained on just layers 38–43. Small file, surprisingly noticeable effect.
The slider is pretty straightforward:
Negative values → very shallow depth of field
Positive values → deep focus / everything sharp
Around -4 to +4 is where I usually work, but it can respond well beyond that. Values around ±6 can produce stronger effects depending on the model, prompt, and how extreme you want the result.
On the negative side, it can push the image toward a very shallow DOF with strong background and foreground separation, giving you some really nice cinematic bokeh. It also works surprisingly well for still images, not just video.
On the positive side, the focus opens up and more of the scene stays sharp, eventually reaching a much more deep-focus / everything-in-focus look.
The fun part is that it doesn't need a huge LoRA to do it. Just a rank 1, ~650 KB file trained on a small set of layers.
Negative = shallow DOF / strong bokeh
0 = natural focus
Positive = deep DOF / everything sharp
As always, the exact sweet spot depends on the model and prompt, so the values are more of a starting point than a hard rule.
SHA256
6980EBDEA60AA78CD05374C7E65590E7DFB177ED8965CDF18F66D841743E9E5B
ID Model ModelTrends.ai
#27292
Laporkan kandungan
Contoh render
Prompt dan tetapan yang dikongsi oleh orang yang membuat render ini di Civitai. Pilih render untuk melihat cara ia dibuat.
Tetapan
Saiz
832x1216
Prompt
IMAGE: A cinematic candid photograph of a 25-year-old ginger woman making fresh bread inside a rustic wooden house during beautiful golden hour. She has natural curly ginger hair, expressive features, and a relaxed, genuine expression. She wears a simple light-colored summer dress, with flour naturally covering her hands, fingers, forearms, parts of her dress, and a few strands of her hair.
She is standing at a rustic wooden table, casually kneading fresh bread dough with both hands. Flour is scattered naturally across the table. A wooden bowl, rolling pin, freshly shaped dough, and a few simple baking utensils are nearby.
Her face, eyes, ginger hair, hands, and upper body are in crisp focus, with natural detailed skin and individual strands of hair visible.
EXTREMELY SHALLOW DEPTH OF FIELD. The foreground is heavily blurred, with the closest edge of the wooden table and scattered flour dissolving into soft shapes. The entire background is strongly defocused into beautiful creamy circular bokeh. The wooden walls, shelves, window, baking utensils, and furniture are recognizable only as soft blurred shapes.
Warm golden-hour sunlight streams through a window behind and slightly beside her, creating large glowing golden bokeh circles and soft warm highlights around her ginger hair and shoulders. The sunlight is beautiful, natural, warm, and cinematic.
Very realistic photography, candid moment, natural imperfections, authentic rustic kitchen, warm golden-hour atmosphere, realistic skin texture, realistic hair, realistic flour, natural pose, subtle depth and dimensionality, soft foreground blur, strong background bokeh, shallow aperture, cinematic photographic look.
A clean horizontal interface labeled "Focus Slider" is positioned along the bottom of the image. The slider knob is at the extreme left, labeled "Shallow DOF".
16:9 composition, camera locked at eye level, natural photographic perspective.
Video: Start on the first frame (shallow DOF, warm golden bokeh, focus slider/knob fully left). A red-haired woman in a white dress kneads a long loaf of dough on a flour-dusted wooden table in a rustic kitchen. Soft natural light streams through the window behind her. Birds sing outside. As she presses and rolls the dough with both flour-dusted hands, the wooden table subtly shifts and creaks under the pressure. Simultaneously the focus slider/knob slowly glides from left to right while depth of field smoothly increases from shallow to deep, gradually revealing sharp detail in the background shelves, window, jars, plants, and hanging lights. She continues the natural kneading motion the entire time. End exactly on the second frame (deep DOF, focus slider/knob fully right).
Candid, natural kitchen sounds: soft dough pressing and folding, light wooden table creaks and shifts, quiet ambient room tone, faint distant outdoor sounds through the open window. Underneath, a low increasing continuous “woummm” sound of the focus slider/knob sliding. No music, no voice-over. Smooth continuous camera, photorealistic, cinematic lighting.
Tetapan
Saiz
1920x960
Prompt
A realistic candid handheld video in a city square. A french ginger 50 years old woman wearing a red jacket, a white shirt and black shoes, sits on a bench eating a small snack while a pigeon stands directly in front of her staring intensely at the food. She slowly moves the snack away, and the pigeon follows it with its eyes. She says: <d>[French] "Tu le veux, bête pigeon?" </d>
She moves it left, then right, and the pigeon tracks every movement with exaggerated seriousness. She eventually gives the pigeon a tiny piece and it immediately walks away as if nothing happened. Natural daylight, realistic bird movement, subtle handheld camera shake, deadpan comedy, ordinary everyday environment.
Tetapan
Saiz
1920x960
Prompt
integrated_multimodal_description: 0.00-1.20s: A whimsical green alien astronaut rides on the back of an orange enormous fantastical space cat flying through deep space. The alien holds onto the cat with one hand while the cat glides gracefully between stars and colorful nebulae. A brilliant rainbow trail streams continuously behind them. The camera follows from a dynamic three-quarter angle. In the distant background, glowing cosmic lettering displays the exact words "Focus Slider".
1.20-2.50s: The space cat suddenly accelerates, weaving playfully around floating asteroids. The alien leans into the turn while the cat's paws stretch forward as if swimming through weightless space. The rainbow trail curves dramatically behind them, remaining continuous and vivid.
2.50-3.50s: The cat makes a joyful leap through a luminous nebula, sending sparkling particles outward as the alien raises one arm triumphantly. The camera pulls back slightly to reveal the full rainbow trail and the surrounding galaxy, with "Focus Slider" still clearly readable in the cosmic background.
overall_soundscape: Subtle futuristic rushing ambience, soft sparkling particle sounds, and playful cat-like vocalizations blended with a spacious cosmic atmosphere.
non_diegetic_music: Whimsical energetic synth music with bright arpeggios, playful percussion, and a magical rising flourish during the final leap.
Tetapan
Saiz
1920x960
Prompt
integrated_multimodal_description: 0.00-2.00s: A 25-year-old blonde woman wearing a flowing red dress holds a pink IPHONE cellphone close to her face while recording herself on a beach. The shot is a handheld selfie perspective, framing her face and upper body as she smiles naturally into the phone camera. Ocean waves and sky are visible behind her. She says: <d>[Russian]"Смотри, что ты теряешь, глупенький!" </d>lowers the cellphone while continuing to smile, then places it carefully upright on the sand with the camera facing toward the beach and ocean. The screen face the viewer where we can see her moving.
2.00-3.50s: From the cellphone's low position on the sand, the woman steps backward into the frame, turns toward the ocean, and begins running across the beach. Her red dress moves dynamically in the sea breeze as her blonde hair flows behind her. The cellphone remains stationary on the sand, creating a stable low-angle view of her movement toward the water.
3.50-5.50s: She runs faster toward the shoreline, reaches the shallow water, and joyfully jumps forward into the incoming wave. Water splashes around her legs and the red dress moves with the impact. The camera remains completely stationary on the sand, capturing her energetic jump against the bright ocean horizon.
5.50-6.00s: She lands in the shallow water with a playful splash, briefly turning her body toward the cellphone as another small wave rolls around her.
overall_soundscape: Close handheld recording sounds at the beginning, followed by soft beach ambience, ocean waves, wind moving through the woman's hair and dress, footsteps on dry sand, then increasingly prominent splashing water as she enters the ocean.
non_diegetic_music: Upbeat, carefree instrumental music with light percussion and bright acoustic elements, beginning softly during the selfie recording and gradually becoming more energetic as she runs toward the water.
Tetapan
Saiz
1920x960
Prompt
integrated_multimodal_description: 0.00-1.20s: A young blond woman rides a skateboard rapidly along a smooth urban promenade, leaning naturally into the movement with confident balance. Her hair and clothing move with the wind. The camera tracks alongside her at low height, keeping her centered while the pavement flows beneath her. She wears a white plain t-shirt and large blue jeans with red tennis. A large colorful street mural behind her displays the exact words "Focus Slider".
1.20-2.50s: She performs a smooth skateboard trick, briefly lifting the board from the pavement and landing cleanly while maintaining her momentum. The camera makes a subtle lateral tracking movement to follow the trick.
2.50-3.50s: She accelerates forward and looks toward the camera with a playful smile as the skateboard rolls rapidly away. The words "Focus Slider" remain clearly visible in the background.
overall_soundscape: Rolling skateboard wheels, subtle pavement vibration, light wind, and distant urban ambience.
non_diegetic_music: Energetic upbeat electronic music with crisp percussion and a short rhythmic accent synchronized with the skateboard trick.
Tetapan
Saiz
1920x960
Prompt
integrated_multimodal_description:
0.00-2.00s: A charming anthropomorphic dog wearing a clean white laboratory coat and black framed square protective glasses sits behind a laboratory workbench. Several transparent glass beakers, test tubes, flasks, and small glass containers filled with distinctly colored chemical substances are arranged across the table. The dog looks seriously focused like a scientist conducting an important experiment. In the background, a large matte-black laboratory chalkboard prominently displays the exact words "Focus Slider" in large white chalk lettering. The dog says:<d>[Korean]"내가 여기서 무슨 짓을 하고 있는지 모르겠어"</d>
2.00s: The dog carefully examines the different chemical substances, shifting its gaze from one glass container to another. It adjusts its glasses with one paw and leans forward with exaggerated scientific concentration. Tiny bubbles rise inside one of the liquids while another container emits a faint wisp of harmless vapor.
3.00s: The dog suddenly looks directly toward the camera with an intensely serious scientific expression, then slowly raises one paw as if it has just discovered an important breakthrough. The laboratory remains visually consistent, with the large blackboard and the words "Focus Slider" behind it.
overall_soundscape: Quiet laboratory room tone, subtle glass clinks, faint bubbling from the chemical containers, occasional soft vapor hiss, and the gentle movement of the dog's laboratory coat.
non_diegetic_music: Playful quirky laboratory-themed instrumental music with light pizzicato strings, subtle percussion, and a gradually building comedic sense of scientific discovery.
Tetapan
Saiz
1920x960
Prompt
A candid handheld video of a 28-year-old man sitting at a small wooden café table on a busy outdoor sidewalk. He has messy short dark brown hair, light stubble, expressive brown eyes, and is wearing a slightly oversized olive-green cotton jacket over a plain cream T-shirt, faded blue jeans, and worn white sneakers. He is laughing while telling a story to his friend when his right elbow accidentally knocks over a white ceramic coffee cup. His eyes suddenly widen in surprise and he reaches forward with both hands trying to catch it, while his friend starts laughing uncontrollably. The table has a small glass of water, a half-eaten croissant, a folded paper napkin, a smartphone, and a metal teaspoon. Behind them are several café tables, colorful umbrellas, potted plants, pedestrians walking along the sidewalk, parked bicycles, and a waiter carrying two plates. Warm late-afternoon sunlight, natural skin texture, spontaneous expressions, realistic hand movements, casual imperfect handheld camera movement, authentic candid atmosphere.
Model serupa
Model AI untuk penjanaan imej, video dan teks
focusslider h3 ialah LoRA untuk keluarga MiniMax H3 yang disenaraikan di ModelTrends.ai, katalog baca sahaja untuk model AI imej, video dan teks sumber terbuka. Bandingkannya dengan model MiniMax H3 lain, semak skor habanya untuk melihat trendnya, dan bukanya di PirateDiffusion atau BitVector untuk mencubanya.
Layari mengikut keluarga model
Model Text / LLM
411
Model Anima
509
Model Chroma
14
Model Flux
1,071
Model Flux 2 / Klein
1,237
Model MiniMax H3
331
Model Hunyuan
197
Model Ideogram
20
Model Krea2
674
Model Qwen
17
Model Qwen2
36
Model Ltx2
150
Model Stable Diffusion 1.5
6,465
Model Stable Diffusion XL 1.0
15,432
Model Zimage
804
Model Wan
919

Model
Trends
.ai
© ModelTrends.ai
|
Dibuat di Jepun
|
© 2026
