Invite vérifiée
Ultra-cinematic 15-second multi-angle action sequence in a dark city street at night, wet asphalt reflecting streetlights, a massive werewolf rampaging through civilians. 0–2s: Wide establishing shot, empty street with a few pedestrians, distant fog and flickering streetlights, subtle tension. 2–3.5s: Close-up on the werewolf emerging from shadows, glowing eyes, low growl, slight camera push-in. 3.5–5s: Hard cut to handheld medium shot as the creature charges forward, heavy footsteps splashing water, people reacting and turning to run. 5–6.5s: Side tracking shot, the werewolf swings its arm in a wide arc, making powerful kinetic contact — a person is launched into the air across frame. 6.5–8s: Slow-motion shot, body spinning mid-air above wet pavement, reflections stretching, debris and droplets suspended in air. 8–10s: Low angle shot, the creature advancing aggressively, another impact sends multiple people sliding and lifting off the ground from force. 10–11.5s: Overhead drone-style shot, crowd scattering in all directions across the street, cars stopped, doors open, chaos spreading. 11.5–13s: Fast whip pan to close-up impact, strong contact motion, particles, water spray, and debris bursting outward. 13–15s: Dynamic tracking backward shot, the werewolf continues advancing toward camera, breathing heavy, final aggressive step as people flee in the background. Environment: wet reflective asphalt, streetlights casting long reflections, light fog, scattered trash and objects reacting to motion. Style: ultra-cinematic realism, high-impact action editing, dramatic cuts, slow-motion accents, heavy motion blur, volumetric atmosphere, no text, no overlays, stable proportions.
Ultra-cinematic 15-second multi-angle action sequence in a dark city street at night, wet asphalt reflecting streetlights, a massive werewolf rampaging through civilians. 0–2s: Wide establishing shot, empty street with a few pedestrians, distant fog and flickering streetlights, subtle tension. 2–3.5s: Close-up on the werewolf emerging from shadows, glowing eyes, low growl, slight camera push-in. 3.5–5s: Hard cut to handheld medium shot as the creature charges forward, heavy footsteps splashing water, people reacting and turning to run. 5–6.5s: Side tracking shot, the werewolf swings its arm in a wide arc, making powerful kinetic contact — a person is launched into the air across frame. 6.5–8s: Slow-motion shot, body spinning mid-air above wet pavement, reflections stretching, debris and droplets suspended in air. 8–10s: Low angle shot, the creature advancing aggressively, another impact sends multiple people sliding and lifting off the ground from force. 10–11.5s: Overhead drone-style shot, crowd scattering in all directions across the street, cars stopped, doors open, chaos spreading. 11.5–13s: Fast whip pan to close-up impact, strong contact motion, particles, water spray, and debris bursting outward. 13–15s: Dynamic tracking backward shot, the werewolf continues advancing toward camera, breathing heavy, final aggressive step as people flee in the background. Environment: wet reflective asphalt, streetlights casting long reflections, light fog, scattered trash and objects reacting to motion. Style: ultra-cinematic realism, high-impact action editing, dramatic cuts, slow-motion accents, heavy motion blur, volumetric atmosphere, no text, no overlays, stable proportions.
Invite vérifiée
@ Image1 is the lead performer — preserve her exact identity in every shot: cornrow braids, septum ring, statement earrings, sculptural glossy white wrapped top, dark indigo denim. @ Audio1 is the music track — entire edit cut to it; no invented music, no voiceover, no lip-sync. Music-video route: dance/performance, brutalist stage show. Track: industrial hip-hop, 95 BPM half-time, distorted 808s, luxury-noir. Director thesis: architecture as choreography — dancers staged on concrete plinths at different heights, a human sculpture garden that activates around the lead. Visual world: brutalist concrete atrium with massive raw concrete plinths and steps at staggered heights, shafts of cold daylight from a skylight, dust motes in the light shafts, dark shadow pockets. Background dancers: 6 dancers in oatmeal and bone-toned knitwear and tailoring, sculptural silhouettes, like museum figures. Palette: bone, oatmeal, concrete grey, deep indigo accent on the lead. No neon, no particles beyond natural dust. Cut map synced to @ Audio1: 0.0–2.0s THE BREATH — static ultra-wide symmetrical: six dancers frozen in sculptural poses on six plinths at different heights, the lead standing on the floor at center like a visitor. Total stillness, museum silence. 2.0s THE DROP — 808 kick: every statue snaps into a new pose simultaneously on the hit, heads turning to her; hard cut to a fast push-in on the lead. 2.0–4.0s she struts between the plinths, camera tracking laterally through foreground columns creating parallax wipes; on each kick one statue above her snaps a new angular pose. 4.0s snare — whip-pan: full freeze, she hits a profile vogue arm frame between two plinths, statues frozen mid-rotation above. 4.0–8.0s sculpture canon: her sharp head isolations and New Way vogue lines on the snares trigger cascading pose-changes across the plinths, a mechanical wave of bodies jumping pose to pose like stop-motion statues; torso rolls on the 808 slides travel up the plinth heights from low to high. Camera slowly cranes upward through the atrium. 8.0s hi-hat triple roll — 3 rapid macro inserts: her earring in a light shaft, a dancer's hand rotating on a plinth edge, dust exploding off concrete as a boot lands. 8.5–12.0s escalation: accelerating orbital-crane move spiraling around the whole sculpture garden as all six dancers now move continuously in their own orbits on the plinths, her arm work at the center speeding up; speed-ramp stretch then snap back on the kick. 12.0s bass slide — the dancers step down off the plinths in unison and sink into deep lunges around her as she glides through a sideways level drop, camera dutch and low across the floor. 13.5–15.0s FINAL BUTTON — hard cut to the opening ultra-wide: she now stands ON the tallest central plinth in a frozen vogue pose one arm high, the six dancers frozen as statues on the floor below looking up; micro push-in, hold. Loopable. Performance rules: lead stoic, dominant; dancers precise, sculptural, expressionless. Continuity: same seven people, same wardrobe, same atrium. Audio intent: every cut and pose-snap on a hit of @ Audio1; faint foley only (steps, fabric, room tone). Quality bar: expensive editorial music video, museum-grade art direction, no AI gloss, no glow.
@ Image1 is the lead performer — preserve her exact identity in every shot: cornrow braids, septum ring, statement earrings, sculptural glossy white wrapped top, dark indigo denim. @ Audio1 is the music track — entire edit cut to it; no invented music, no voiceover, no lip-sync. Music-video route: dance/performance, brutalist stage show. Track: industrial hip-hop, 95 BPM half-time, distorted 808s, luxury-noir. Director thesis: architecture as choreography — dancers staged on concrete plinths at different heights, a human sculpture garden that activates around the lead. Visual world: brutalist concrete atrium with massive raw concrete plinths and steps at staggered heights, shafts of cold daylight from a skylight, dust motes in the light shafts, dark shadow pockets. Background dancers: 6 dancers in oatmeal and bone-toned knitwear and tailoring, sculptural silhouettes, like museum figures. Palette: bone, oatmeal, concrete grey, deep indigo accent on the lead. No neon, no particles beyond natural dust. Cut map synced to @ Audio1: 0.0–2.0s THE BREATH — static ultra-wide symmetrical: six dancers frozen in sculptural poses on six plinths at different heights, the lead standing on the floor at center like a visitor. Total stillness, museum silence. 2.0s THE DROP — 808 kick: every statue snaps into a new pose simultaneously on the hit, heads turning to her; hard cut to a fast push-in on the lead. 2.0–4.0s she struts between the plinths, camera tracking laterally through foreground columns creating parallax wipes; on each kick one statue above her snaps a new angular pose. 4.0s snare — whip-pan: full freeze, she hits a profile vogue arm frame between two plinths, statues frozen mid-rotation above. 4.0–8.0s sculpture canon: her sharp head isolations and New Way vogue lines on the snares trigger cascading pose-changes across the plinths, a mechanical wave of bodies jumping pose to pose like stop-motion statues; torso rolls on the 808 slides travel up the plinth heights from low to high. Camera slowly cranes upward through the atrium. 8.0s hi-hat triple roll — 3 rapid macro inserts: her earring in a light shaft, a dancer's hand rotating on a plinth edge, dust exploding off concrete as a boot lands. 8.5–12.0s escalation: accelerating orbital-crane move spiraling around the whole sculpture garden as all six dancers now move continuously in their own orbits on the plinths, her arm work at the center speeding up; speed-ramp stretch then snap back on the kick. 12.0s bass slide — the dancers step down off the plinths in unison and sink into deep lunges around her as she glides through a sideways level drop, camera dutch and low across the floor. 13.5–15.0s FINAL BUTTON — hard cut to the opening ultra-wide: she now stands ON the tallest central plinth in a frozen vogue pose one arm high, the six dancers frozen as statues on the floor below looking up; micro push-in, hold. Loopable. Performance rules: lead stoic, dominant; dancers precise, sculptural, expressionless. Continuity: same seven people, same wardrobe, same atrium. Audio intent: every cut and pose-snap on a hit of @ Audio1; faint foley only (steps, fabric, room tone). Quality bar: expensive editorial music video, museum-grade art direction, no AI gloss, no glow.
Invite vérifiée
Preserve the exact face, identity, skin tone, body proportions, facial features, hairstyle, and outfit from @image1 throughout the entire video. Do not change her appearance in any scene. Duration: 15 seconds Aspect Ratio: 16:9 Style: Authentic Japanese 2026 daily-life vlog. Extremely realistic handheld smartphone front-camera footage. Raw phone video aesthetic with natural camera shake, autofocus hunting, slight exposure fluctuations, occasional imperfect framing, natural motion blur, genuine smartphone microphone audio, no cinematic lighting, no beauty filter, no stabilization, no color grading, no subtitles, no watermark. 00:00–00:02 Walking through a quiet modern Japanese residential neighborhood while filming herself. Several local mothers are chatting outside a house. She smiles warmly, waves, and naturally says, "Ohayō gozaimasu!" The mothers smile and reply, "Ohayō gozaimasu!" 00:02–00:04 She stops at a small neighborhood convenience store and buys a cup of iced tea. The cashier hands it to her. She smiles and says, "Arigatō gozaimasu." She immediately takes a sip while continuing her vlog. 00:04–00:06 She notices a local bus about to leave. Still holding her iced tea, she jogs toward it while laughing, raises one hand, and calls, "Sumimasen! Matte kudasai!" The driver slows down and opens the door. 00:06–00:08 She steps onto the bus, laughing because she barely made it. She briefly films the passengers and the busy street outside through the window. 00:08–00:10 After getting off, she notices a street musician playing an acoustic guitar. She smiles, places some coins into the donation box, and says, "Ganbatte kudasai!" 00:10–00:12 Continuing her walk along a lively Japanese shopping street, casually drinking her iced tea while chatting naturally to the camera about enjoying simple moments around the city. 00:12–00:15 Selfie shot while walking away. She smiles brightly, waves at the camera, and says, "Mata ne! Arigatō gozaimashita!" The recording ends naturally as the phone lowers slightly. Natural ambience only: Neighborhood conversations, bicycles passing, distant traffic, bus engine sounds, footsteps, ice clinking inside the cup, convenience store ambience, street musician singing softly with acoustic guitar, light breeze, authentic Japanese city ambience.
Preserve the exact face, identity, skin tone, body proportions, facial features, hairstyle, and outfit from @image1 throughout the entire video. Do not change her appearance in any scene. Duration: 15 seconds Aspect Ratio: 16:9 Style: Authentic Japanese 2026 daily-life vlog. Extremely realistic handheld smartphone front-camera footage. Raw phone video aesthetic with natural camera shake, autofocus hunting, slight exposure fluctuations, occasional imperfect framing, natural motion blur, genuine smartphone microphone audio, no cinematic lighting, no beauty filter, no stabilization, no color grading, no subtitles, no watermark. 00:00–00:02 Walking through a quiet modern Japanese residential neighborhood while filming herself. Several local mothers are chatting outside a house. She smiles warmly, waves, and naturally says, "Ohayō gozaimasu!" The mothers smile and reply, "Ohayō gozaimasu!" 00:02–00:04 She stops at a small neighborhood convenience store and buys a cup of iced tea. The cashier hands it to her. She smiles and says, "Arigatō gozaimasu." She immediately takes a sip while continuing her vlog. 00:04–00:06 She notices a local bus about to leave. Still holding her iced tea, she jogs toward it while laughing, raises one hand, and calls, "Sumimasen! Matte kudasai!" The driver slows down and opens the door. 00:06–00:08 She steps onto the bus, laughing because she barely made it. She briefly films the passengers and the busy street outside through the window. 00:08–00:10 After getting off, she notices a street musician playing an acoustic guitar. She smiles, places some coins into the donation box, and says, "Ganbatte kudasai!" 00:10–00:12 Continuing her walk along a lively Japanese shopping street, casually drinking her iced tea while chatting naturally to the camera about enjoying simple moments around the city. 00:12–00:15 Selfie shot while walking away. She smiles brightly, waves at the camera, and says, "Mata ne! Arigatō gozaimashita!" The recording ends naturally as the phone lowers slightly. Natural ambience only: Neighborhood conversations, bicycles passing, distant traffic, bus engine sounds, footsteps, ice clinking inside the cup, convenience store ambience, street musician singing softly with acoustic guitar, light breeze, authentic Japanese city ambience.
Invite vérifiée
Create a 15-second ultra-realistic cinematic historical time-travel selfie vlog about **[HISTORICAL MOMENT]**. The protagonist is a naturally beautiful South Asian woman in her mid-20s with warm brown skin, expressive dark brown eyes, long flowing black hair, soft symmetrical facial features, and a graceful, athletic build. Maintain the exact same face, hairstyle, body proportions, skin tone, and identity throughout the entire video. She is an ordinary traveler who has accidentally stepped into history and is documenting the experience in real time. She wears historically accurate everyday clothing appropriate to the exact era and location. Her outfit may naturally become dusty, muddy, smoky, wet, or worn as events unfold, but it must always remain period-accurate. Never include modern clothing, logos, jewelry, watches, sunglasses, headphones, backpacks, or any modern objects. The entire video is filmed as one continuous handheld selfie shot. The camera is always held at arm's length with realistic handheld movement, subtle walking shake, natural autofocus, slight wide-angle smartphone perspective, realistic motion blur, cinematic lighting, and documentary-quality realism. The phone itself is never visible. Never switch to a third-person view. The video begins with the woman suddenly arriving in **[HISTORICAL MOMENT]**, looking around in amazement before speaking directly to the camera. As she walks through the environment, she reacts naturally, points toward important events happening behind her, briefly examines historically accurate objects or locations, and shares her thoughts like a modern travel vlogger. She naturally passes by the most iconic historical figure or witnesses the defining historical event without changing history or interacting in a way that alters the timeline. As the event reaches its climax, she reacts with genuine awe before ending with a powerful selfie that perfectly frames the most iconic landmark or historical moment in the background. Everything must be historically accurate, including architecture, clothing, hairstyles, tools, weapons, transportation, technology, streets, weather, lighting, landscapes, crowds, animals, customs, social behavior, and environmental details appropriate to the exact year and location. The world should feel alive with authentic movement, smoke, dust, fire, rain, wind, mud, and realistic crowd activity. The woman speaks naturally in English throughout the video with authentic emotion, curiosity, excitement, and wonder. Include synchronized **burned-in English subtitles** at the bottom of the screen that exactly match her spoken dialogue in a clean cinematic font. Style: ultra-photorealistic, cinematic documentary, immersive, emotionally engaging, natural human expressions, realistic physics, high-detail textures, professional color grading, 8K quality, indistinguishable from real smartphone footage captured by someone who accidentally traveled back in time. No CGI look, no fantasy elements, no anachronisms, and no historical inaccuracies. For the The sinking of titanic
Create a 15-second ultra-realistic cinematic historical time-travel selfie vlog about **[HISTORICAL MOMENT]**. The protagonist is a naturally beautiful South Asian woman in her mid-20s with warm brown skin, expressive dark brown eyes, long flowing black hair, soft symmetrical facial features, and a graceful, athletic build. Maintain the exact same face, hairstyle, body proportions, skin tone, and identity throughout the entire video. She is an ordinary traveler who has accidentally stepped into history and is documenting the experience in real time. She wears historically accurate everyday clothing appropriate to the exact era and location. Her outfit may naturally become dusty, muddy, smoky, wet, or worn as events unfold, but it must always remain period-accurate. Never include modern clothing, logos, jewelry, watches, sunglasses, headphones, backpacks, or any modern objects. The entire video is filmed as one continuous handheld selfie shot. The camera is always held at arm's length with realistic handheld movement, subtle walking shake, natural autofocus, slight wide-angle smartphone perspective, realistic motion blur, cinematic lighting, and documentary-quality realism. The phone itself is never visible. Never switch to a third-person view. The video begins with the woman suddenly arriving in **[HISTORICAL MOMENT]**, looking around in amazement before speaking directly to the camera. As she walks through the environment, she reacts naturally, points toward important events happening behind her, briefly examines historically accurate objects or locations, and shares her thoughts like a modern travel vlogger. She naturally passes by the most iconic historical figure or witnesses the defining historical event without changing history or interacting in a way that alters the timeline. As the event reaches its climax, she reacts with genuine awe before ending with a powerful selfie that perfectly frames the most iconic landmark or historical moment in the background. Everything must be historically accurate, including architecture, clothing, hairstyles, tools, weapons, transportation, technology, streets, weather, lighting, landscapes, crowds, animals, customs, social behavior, and environmental details appropriate to the exact year and location. The world should feel alive with authentic movement, smoke, dust, fire, rain, wind, mud, and realistic crowd activity. The woman speaks naturally in English throughout the video with authentic emotion, curiosity, excitement, and wonder. Include synchronized **burned-in English subtitles** at the bottom of the screen that exactly match her spoken dialogue in a clean cinematic font. Style: ultra-photorealistic, cinematic documentary, immersive, emotionally engaging, natural human expressions, realistic physics, high-detail textures, professional color grading, 8K quality, indistinguishable from real smartphone footage captured by someone who accidentally traveled back in time. No CGI look, no fantasy elements, no anachronisms, and no historical inaccuracies. For the The sinking of titanic
Invite vérifiée
Format: 16:9 Concept: "Late-Night Practice Room Diary" The entire video is filmed by the idol herself in a genuine first-person vlog style using a compact early-2000s digital camcorder. The camera is never visible. The footage should feel spontaneous and personal rather than produced, with subtle handheld movement, realistic autofocus adjustments, slight exposure changes when moving between brighter and darker areas, natural walking motion, and authentic framing imperfections. Everything unfolds as one uninterrupted recording with no cuts or transitions. Visual Style Authentic MiniDV aesthetic with soft detail, light digital grain, muted contrast, accurate skin tones, slight highlight bloom under fluorescent lights, and subtle tape compression. No cinematic grading, beauty filters, exaggerated depth of field, or artificial visual effects. Character SOOJIN, a Korean female idol in her twenties with long naturally straight black hair, fresh dewy skin, expressive eyes, and understated stage makeup. She wears loose rehearsal clothes—an oversized sweatshirt, comfortable joggers, sneakers, and a practice badge clipped to her hoodie. Her personality is relaxed, playful, and naturally expressive. Location A quiet rehearsal floor inside an entertainment company building late in the evening. Long mirrored corridors, dance studios with glass windows, stacked speakers, folded chairs, water dispensers, vending machines, rehearsal schedules pinned to the walls, and occasional dancers or staff passing naturally in the background. The recording begins as SOOJIN quietly opens the door to an empty hallway outside a dance studio. She smiles and whispers, "Everyone's still practicing... let's see who's left." She walks down the corridor, occasionally aiming the camera through studio windows where blurred silhouettes continue rehearsing in the background. Stopping beside a vending machine, she laughs. "I've walked past this thing at least twenty times today." She buys a drink, catches it, and takes a sip while continuing through the hallway. Further ahead, she finds a wall covered with handwritten encouragement notes from different artists. She slowly pans across them before adding her own short message with a marker. She grins at the camera. "Now I have to debut again just to read this later." As she continues walking, distant music suddenly stops throughout the building. The unexpected silence makes her pause for a second before smiling. "I guess practice is finally over." She reaches a large window overlooking Seoul at night, resting the camera on the ledge for a brief steady moment while city lights fill the background. Without speaking, she simply enjoys the view for several seconds. A manager's voice echoes from somewhere down the hallway, calling everyone together. She immediately picks the camera back up. "That's my signal." Still smiling, she walks toward the elevator with relaxed confidence, gives one final wave to the camera as the elevator doors open, steps inside, and the recording naturally ends as the doors close. Generation Requirements Produce one continuous, uninterrupted recording with realistic body movement, stable facial identity, physically accurate interactions, believable lighting transitions, natural ambient audio, consistent proportions, no duplicated people or objects, no anatomical errors, no AI artifacts, no jump cuts, and no unrealistic camera behavior. The result should feel indistinguishable from a real personal backstage vlog captured by an idol after rehearsal.
Format: 16:9 Concept: "Late-Night Practice Room Diary" The entire video is filmed by the idol herself in a genuine first-person vlog style using a compact early-2000s digital camcorder. The camera is never visible. The footage should feel spontaneous and personal rather than produced, with subtle handheld movement, realistic autofocus adjustments, slight exposure changes when moving between brighter and darker areas, natural walking motion, and authentic framing imperfections. Everything unfolds as one uninterrupted recording with no cuts or transitions. Visual Style Authentic MiniDV aesthetic with soft detail, light digital grain, muted contrast, accurate skin tones, slight highlight bloom under fluorescent lights, and subtle tape compression. No cinematic grading, beauty filters, exaggerated depth of field, or artificial visual effects. Character SOOJIN, a Korean female idol in her twenties with long naturally straight black hair, fresh dewy skin, expressive eyes, and understated stage makeup. She wears loose rehearsal clothes—an oversized sweatshirt, comfortable joggers, sneakers, and a practice badge clipped to her hoodie. Her personality is relaxed, playful, and naturally expressive. Location A quiet rehearsal floor inside an entertainment company building late in the evening. Long mirrored corridors, dance studios with glass windows, stacked speakers, folded chairs, water dispensers, vending machines, rehearsal schedules pinned to the walls, and occasional dancers or staff passing naturally in the background. The recording begins as SOOJIN quietly opens the door to an empty hallway outside a dance studio. She smiles and whispers, "Everyone's still practicing... let's see who's left." She walks down the corridor, occasionally aiming the camera through studio windows where blurred silhouettes continue rehearsing in the background. Stopping beside a vending machine, she laughs. "I've walked past this thing at least twenty times today." She buys a drink, catches it, and takes a sip while continuing through the hallway. Further ahead, she finds a wall covered with handwritten encouragement notes from different artists. She slowly pans across them before adding her own short message with a marker. She grins at the camera. "Now I have to debut again just to read this later." As she continues walking, distant music suddenly stops throughout the building. The unexpected silence makes her pause for a second before smiling. "I guess practice is finally over." She reaches a large window overlooking Seoul at night, resting the camera on the ledge for a brief steady moment while city lights fill the background. Without speaking, she simply enjoys the view for several seconds. A manager's voice echoes from somewhere down the hallway, calling everyone together. She immediately picks the camera back up. "That's my signal." Still smiling, she walks toward the elevator with relaxed confidence, gives one final wave to the camera as the elevator doors open, steps inside, and the recording naturally ends as the doors close. Generation Requirements Produce one continuous, uninterrupted recording with realistic body movement, stable facial identity, physically accurate interactions, believable lighting transitions, natural ambient audio, consistent proportions, no duplicated people or objects, no anatomical errors, no AI artifacts, no jump cuts, and no unrealistic camera behavior. The result should feel indistinguishable from a real personal backstage vlog captured by an idol after rehearsal.
Invite vérifiée
经典的90年代迪士尼动画风格。场景一:在阴暗的埃及地牢里,强大的胡狼神阿努比斯被锁链束缚在墙上,双臂高举过头,腋窝暴露在外。他神情骄傲而桀骜不驯,两名成年审讯员站在一旁。其中一名审讯员走近,厉声喝道:“说话,阿努比斯!”场景二:阿努比斯拒绝回答,默默地怒视着审讯员。审讯员嘴角勾起一丝邪笑,一只手伸进阿努比斯裸露的腋窝,用轻柔流畅的手指轻轻地挠着。审讯员戏谑地说:“咯咯咯。”场景三:阿努比斯严肃的表情瞬间转为惊讶。他闭上眼睛,咧嘴一笑,大笑起来。他扭动挣扎,双臂仍被锁链高举过头,腋窝暴露在外。场景4:审讯者继续挠阿努比斯的腋窝,而不是肋骨、胸膛或侧腰。阿努比斯笑得更大声了,咯咯地笑着说:“停!我好怕痒!” 挠痒痒一直持续到阿努比斯终于脱口而出:“好吧!我说话!” 然后又开始大笑起来。阿努比斯是一位肌肉发达的黑色拟人化胡狼神,手臂和脖子上戴着金色的环。他的声音低沉而威严。配音为英语。
经典的90年代迪士尼动画风格。场景一:在阴暗的埃及地牢里,强大的胡狼神阿努比斯被锁链束缚在墙上,双臂高举过头,腋窝暴露在外。他神情骄傲而桀骜不驯,两名成年审讯员站在一旁。其中一名审讯员走近,厉声喝道:“说话,阿努比斯!”场景二:阿努比斯拒绝回答,默默地怒视着审讯员。审讯员嘴角勾起一丝邪笑,一只手伸进阿努比斯裸露的腋窝,用轻柔流畅的手指轻轻地挠着。审讯员戏谑地说:“咯咯咯。”场景三:阿努比斯严肃的表情瞬间转为惊讶。他闭上眼睛,咧嘴一笑,大笑起来。他扭动挣扎,双臂仍被锁链高举过头,腋窝暴露在外。场景4:审讯者继续挠阿努比斯的腋窝,而不是肋骨、胸膛或侧腰。阿努比斯笑得更大声了,咯咯地笑着说:“停!我好怕痒!” 挠痒痒一直持续到阿努比斯终于脱口而出:“好吧!我说话!” 然后又开始大笑起来。阿努比斯是一位肌肉发达的黑色拟人化胡狼神,手臂和脖子上戴着金色的环。他的声音低沉而威严。配音为英语。
Invite vérifiée
Scene & Mood: Surreal daytime park editorial with calm anti-gravity elegance. A red-haired woman floats horizontally above a bright green lawn as if reclining in invisible air, holding a clear glass in one hand and a green bottle in the other. The mood is stylish, dry, playful, and cinematic, with the impossible levitation treated as completely real inside a sunny city park. Frame Map: Begin in a full-body side-profile composition. image1 floats horizontally across the center of frame, suspended roughly two feet above the grass, head on the left side, legs lifted diagonally toward the upper right. Keep her full body visible with clean space between her shadow and the grass below. The lawn fills the lower foreground, trees and park railing sit in the midground, and the lake with distant city skyline sits in the background. Maintain a medium-wide environmental fashion frame with enough negative blue sky above her raised shoe and enough grass below to show the floating shadow. Subject Lock image1: image1 keeps the same face, hair, wardrobe, body shape, and silhouette throughout. Preserve her red hair, narrow sunglasses, calm facial expression, oversized white button-up shirt, black crossbody strap, black wide-leg trousers, black loafers, clear drinking glass in her raised left hand, and green bottle held near her torso. Keep her reclining floating pose, composed attitude, slim silhouette, and dry editorial expression. Do not change her hair color, sunglasses, shirt volume, trouser shape, shoe style, bottle color, or glass shape. Cross-Frame Rules: Keep image1 as the only person and main subject. Maintain the same levitation height, horizontal body orientation, wardrobe, props, and calm expression for all 15 seconds. The glass stays in her left hand and the green bottle stays in her other hand. No spilling, no dropped props, no duplicated limbs, no body warping, no floating extra objects, no sudden pose reset. The 360 camera orbit may reveal different sides of the park, but the subject remains centered, stable, and fully readable. No on-screen text, no captions, no signage typography, no rendered text in the frame. Movement: 15 seconds total. image1 floats steadily above the grass with a slow, subtle hover drift, as if held by invisible air. Her legs remain lifted and relaxed; one shoe points slightly upward while the other leg bends naturally. Micro-motion: slow breathing, a tiny wrist adjustment holding the glass, slight fingers tightening around the bottle, hair moving gently in the breeze, shirt fabric puffing and settling, trouser hems swaying softly, sunglasses catching small sky reflections. Environmental motion: grass blades ripple, tree leaves move in light wind, lake water shimmers, distant park life stays soft and indistinct, and her shadow glides lightly over the lawn beneath her. Camera motion: one smooth 360-degree orbit around her over the full 15 seconds, keeping her centered and full body visible, with a controlled low-to-mid camera height that circles from her side profile, around her feet, past the opposite side, and back toward the starting angle. Last Frame: At 15 seconds, the camera completes the 360 orbit and settles close to the original side-profile composition. image1 remains floating calmly above the grass, head left, legs raised right, glass lifted with relaxed confidence and green bottle held near her torso. Her shadow sits clearly on the lawn below, trees and skyline softly behind her, expression unchanged behind sunglasses. Camera eases to a steady centered frame with no visible shake. No on-screen text, no captions, no signage typography, no rendered text in the frame. World Plate: Sunny city park in clear daylight with vivid green grass, leafy trees, black park railing, a calm lake, and distant skyline softened by atmospheric depth. Bright blue sky with natural daylight, soft shadows under the floating body, realistic grass texture, mild heat shimmer over the lawn, and gentle background softness. The environment stays believable and grounded, making the levitation feel clean, practical, and photographed. Sound Bed: Diegetic audio only: soft park breeze, grass rustle, faint leaf movement, distant lake ambience, subtle glass liquid movement, tiny bottle-and-glass contact, light fabric flutter, distant indistinct park sounds. No music, no soundtrack, no score, no lyrics. Capture Realism: Natural skin texture, matte skin, realistic red hair movement, real cotton shirt folds, believable trouser weight, grounded shoe reflections, soft highlight rolloff, lifted shadows, mild 35mm grain, realistic floating shadow, clean daylight exposure, and gentle background softness. Avoid AI gloss, plastic skin, over-sharp grass, warped hands, melted props, duplicated body parts, crushed blacks, clipped whites, and fake levitation blur. Camera Capture: 15 seconds, 24fps, 180-degree shutter, wide-latitude cinema capture, smooth stabilized 360-degree orbit with subtle handheld breath, 35mm lens for full-body environmental fashion framing, fine 35mm grain, soft highlight rolloff, natural daylight color grade, realistic motion blur, stable subject tracking, and practical editorial realism.
Scene & Mood: Surreal daytime park editorial with calm anti-gravity elegance. A red-haired woman floats horizontally above a bright green lawn as if reclining in invisible air, holding a clear glass in one hand and a green bottle in the other. The mood is stylish, dry, playful, and cinematic, with the impossible levitation treated as completely real inside a sunny city park. Frame Map: Begin in a full-body side-profile composition. image1 floats horizontally across the center of frame, suspended roughly two feet above the grass, head on the left side, legs lifted diagonally toward the upper right. Keep her full body visible with clean space between her shadow and the grass below. The lawn fills the lower foreground, trees and park railing sit in the midground, and the lake with distant city skyline sits in the background. Maintain a medium-wide environmental fashion frame with enough negative blue sky above her raised shoe and enough grass below to show the floating shadow. Subject Lock image1: image1 keeps the same face, hair, wardrobe, body shape, and silhouette throughout. Preserve her red hair, narrow sunglasses, calm facial expression, oversized white button-up shirt, black crossbody strap, black wide-leg trousers, black loafers, clear drinking glass in her raised left hand, and green bottle held near her torso. Keep her reclining floating pose, composed attitude, slim silhouette, and dry editorial expression. Do not change her hair color, sunglasses, shirt volume, trouser shape, shoe style, bottle color, or glass shape. Cross-Frame Rules: Keep image1 as the only person and main subject. Maintain the same levitation height, horizontal body orientation, wardrobe, props, and calm expression for all 15 seconds. The glass stays in her left hand and the green bottle stays in her other hand. No spilling, no dropped props, no duplicated limbs, no body warping, no floating extra objects, no sudden pose reset. The 360 camera orbit may reveal different sides of the park, but the subject remains centered, stable, and fully readable. No on-screen text, no captions, no signage typography, no rendered text in the frame. Movement: 15 seconds total. image1 floats steadily above the grass with a slow, subtle hover drift, as if held by invisible air. Her legs remain lifted and relaxed; one shoe points slightly upward while the other leg bends naturally. Micro-motion: slow breathing, a tiny wrist adjustment holding the glass, slight fingers tightening around the bottle, hair moving gently in the breeze, shirt fabric puffing and settling, trouser hems swaying softly, sunglasses catching small sky reflections. Environmental motion: grass blades ripple, tree leaves move in light wind, lake water shimmers, distant park life stays soft and indistinct, and her shadow glides lightly over the lawn beneath her. Camera motion: one smooth 360-degree orbit around her over the full 15 seconds, keeping her centered and full body visible, with a controlled low-to-mid camera height that circles from her side profile, around her feet, past the opposite side, and back toward the starting angle. Last Frame: At 15 seconds, the camera completes the 360 orbit and settles close to the original side-profile composition. image1 remains floating calmly above the grass, head left, legs raised right, glass lifted with relaxed confidence and green bottle held near her torso. Her shadow sits clearly on the lawn below, trees and skyline softly behind her, expression unchanged behind sunglasses. Camera eases to a steady centered frame with no visible shake. No on-screen text, no captions, no signage typography, no rendered text in the frame. World Plate: Sunny city park in clear daylight with vivid green grass, leafy trees, black park railing, a calm lake, and distant skyline softened by atmospheric depth. Bright blue sky with natural daylight, soft shadows under the floating body, realistic grass texture, mild heat shimmer over the lawn, and gentle background softness. The environment stays believable and grounded, making the levitation feel clean, practical, and photographed. Sound Bed: Diegetic audio only: soft park breeze, grass rustle, faint leaf movement, distant lake ambience, subtle glass liquid movement, tiny bottle-and-glass contact, light fabric flutter, distant indistinct park sounds. No music, no soundtrack, no score, no lyrics. Capture Realism: Natural skin texture, matte skin, realistic red hair movement, real cotton shirt folds, believable trouser weight, grounded shoe reflections, soft highlight rolloff, lifted shadows, mild 35mm grain, realistic floating shadow, clean daylight exposure, and gentle background softness. Avoid AI gloss, plastic skin, over-sharp grass, warped hands, melted props, duplicated body parts, crushed blacks, clipped whites, and fake levitation blur. Camera Capture: 15 seconds, 24fps, 180-degree shutter, wide-latitude cinema capture, smooth stabilized 360-degree orbit with subtle handheld breath, 35mm lens for full-body environmental fashion framing, fine 35mm grain, soft highlight rolloff, natural daylight color grade, realistic motion blur, stable subject tracking, and practical editorial realism.
Invite vérifiée
京剧非遗短片,电影感,东方美学,温暖克制。传统戏班后台与手工作坊里,老师傅安静制作京剧头饰、整理戏服、勾 勒脸谱,手部细节细腻,丝线、珠饰、颜料和绣纹都很有质感。一个年轻学徒在旁边认真看着,随后小心接过工具,在老师傅指导 下完成细小步骤。老师傅为他扶正头面、整理衣襟,像把一门手艺和一份情感轻轻交到他手里。最后年轻人穿戴整齐,站在即将上 场的戏台边,微光照亮戏服与侧脸,老师傅在身后安静注视。整体氛围安静、深情、有传承感,少字幕,穿插合适的台词。
京剧非遗短片,电影感,东方美学,温暖克制。传统戏班后台与手工作坊里,老师傅安静制作京剧头饰、整理戏服、勾 勒脸谱,手部细节细腻,丝线、珠饰、颜料和绣纹都很有质感。一个年轻学徒在旁边认真看着,随后小心接过工具,在老师傅指导 下完成细小步骤。老师傅为他扶正头面、整理衣襟,像把一门手艺和一份情感轻轻交到他手里。最后年轻人穿戴整齐,站在即将上 场的戏台边,微光照亮戏服与侧脸,老师傅在身后安静注视。整体氛围安静、深情、有传承感,少字幕,穿插合适的台词。
Invite vérifiée
Fisheye lens skatepark shot on a sunny day with chain-link fences and graffiti-covered ramps in the background. Camera starts low on the empty concrete bowl, then a skater in a white t-shirt, baggy jeans, and a red cap rolls into frame - standing on top of a dachshund dog that runs along like a living skateboard, its little legs pumping. The skater balances and rides the dachshund through the park with confident style. Warm summer tones, VHS-style vignette, 60fps action, humorou and surreal.
Fisheye lens skatepark shot on a sunny day with chain-link fences and graffiti-covered ramps in the background. Camera starts low on the empty concrete bowl, then a skater in a white t-shirt, baggy jeans, and a red cap rolls into frame - standing on top of a dachshund dog that runs along like a living skateboard, its little legs pumping. The skater balances and rides the dachshund through the park with confident style. Warm summer tones, VHS-style vignette, 60fps action, humorou and surreal.
Invite vérifiée
A super cute chibi-style anime cartoon little penguin girl with big sparkling eyes, fluffy black and white feathers, tiny wings, and a small pink bow on her head. She's sitting at a table happily eating her favorite food with chopsticks. Suddenly, a finger gently pokes her soft cheek from the side. Show her dramatic adorable reaction in sequence: eyes widening in surprise, cheeks puffing out, intense blush, a funny puffed-up angry-cute expression with tiny steam coming from her head, then she turns and glares cutely at the finger while still chewing. Kawaii anime style, soft pastel colors, smooth 2D animation, vibrant and wholesome, high detail, dynamic short animation sequence, funny and heartwarming mood, clean bright background.
A super cute chibi-style anime cartoon little penguin girl with big sparkling eyes, fluffy black and white feathers, tiny wings, and a small pink bow on her head. She's sitting at a table happily eating her favorite food with chopsticks. Suddenly, a finger gently pokes her soft cheek from the side. Show her dramatic adorable reaction in sequence: eyes widening in surprise, cheeks puffing out, intense blush, a funny puffed-up angry-cute expression with tiny steam coming from her head, then she turns and glares cutely at the finger while still chewing. Kawaii anime style, soft pastel colors, smooth 2D animation, vibrant and wholesome, high detail, dynamic short animation sequence, funny and heartwarming mood, clean bright background.
Invite vérifiée
FPV shot: The camera dives into a collapsing glass tower during sunrise, racing through floors filled with exploding windows, falling steel beams, and floating office debris. Sunlight hits thousands of glass shards spinning in slow motion, creating sharp reflections across the frame. The camera spirals downward as the building implodes around it, dodging crystal fragments and dust clouds. For one perfect second, the shards align with the sun and reveal "SHATTER" before the entire structure collapses into a massive cloud of glass and concrete. Elegant destruction, premium VFX, cinematic action.
FPV shot: The camera dives into a collapsing glass tower during sunrise, racing through floors filled with exploding windows, falling steel beams, and floating office debris. Sunlight hits thousands of glass shards spinning in slow motion, creating sharp reflections across the frame. The camera spirals downward as the building implodes around it, dodging crystal fragments and dust clouds. For one perfect second, the shards align with the sun and reveal "SHATTER" before the entire structure collapses into a massive cloud of glass and concrete. Elegant destruction, premium VFX, cinematic action.
Invite vérifiée
Vertical 9:16 K-pop dance video, The dancer faces camera and performs sharp cute K-pop choreography: small rhythmic knee bounces, step-touch footwork, hips shifting side to side, palms low and open, wrists flick outward, shoulders roll, arms sweep from waist to chest, quick hand accents near the face, point to camera, body wave, then return to center pose. Facial expression is important: start with a cool confident gaze, slight pout, relaxed lips. On the beat, switch to playful smirk, quick eyebrow lift, direct eye contact with the camera. During hand-to-face accents, make a teasing wink or soft half-smile. On the chorus hits, open the mouth slightly as if mouthing the lyric, then finish with a sharp charming smile and confident idol stare. Keep expressions natural, cute, flirty, and synchronized with the choreography. Stable camera with slight push-in, clean realistic motion, smooth full-body dance, natural hands and legs, realistic anatomy.
Vertical 9:16 K-pop dance video, The dancer faces camera and performs sharp cute K-pop choreography: small rhythmic knee bounces, step-touch footwork, hips shifting side to side, palms low and open, wrists flick outward, shoulders roll, arms sweep from waist to chest, quick hand accents near the face, point to camera, body wave, then return to center pose. Facial expression is important: start with a cool confident gaze, slight pout, relaxed lips. On the beat, switch to playful smirk, quick eyebrow lift, direct eye contact with the camera. During hand-to-face accents, make a teasing wink or soft half-smile. On the chorus hits, open the mouth slightly as if mouthing the lyric, then finish with a sharp charming smile and confident idol stare. Keep expressions natural, cute, flirty, and synchronized with the choreography. Stable camera with slight push-in, clean realistic motion, smooth full-body dance, natural hands and legs, realistic anatomy.
Explorez les flux de travail vidéo FLUX 3
Commencez par une scène écrite, donnez vie à une image de référence, ou explorez la Vitrine de prompts pour une direction créative.
Créez une vidéo à partir d'un texte ou d'une image
Utilisez les modèles vidéo FLUX pour créer une courte scène destinée à une publication sociale, un récit produit ou un storyboard. Choisissez le mode d'entrée et les contrôles affichés dans le générateur.
Décrivez votre sujet et son mouvement, choisissez les paramètres pris en charge et vérifiez le coût en crédits. Le statut de votre requête et les résultats terminés sont disponibles dans l'Historique.
Créez votre première vidéo
Élaborez un brief vidéo FLUX ciblé

Écrivez une scène avec une action claire
Décrivez le sujet, le décor, la taille du plan, le mouvement de caméra et l'éclairage. Utilisez l'assistant de prompt lorsque vous souhaitez de l'aide pour développer une idée approximative, puis modifiez la direction que vous préférez.
Explorer les idées de vidéos et les guides créatifs
Utilisez la Prompt Showcase pour étudier les résultats publiés et leur modèle réel. Lisez le blog pour découvrir des méthodes pratiques d'élaboration d'un brief vidéo et d'évaluation de votre résultat.
Un espace de travail pour de courtes histoires visuelles
Partez d'un texte ou d'une image, donnez à votre scène une direction sonore lorsque cela est pris en charge, et choisissez les paramètres de sortie affichés pour le modèle sélectionné.

Les modèles vidéo FLUX dans un seul espace de travail
Le menu des modèles affiche les modèles vidéo FLUX disponibles sur flux3 video. Chaque modèle possède ses propres entrées, paramètres et coûts en crédits. Vérifiez la sélection avant de générer.

Directions sonores pour votre scène
Lorsque l'audio est pris en charge, décrivez séparément les dialogues, les sons d'ambiance et la musique des instructions visuelles. Lisez le clip terminé et vérifiez le son et le minutage avant de publier.

Texte vers vidéo et image vers vidéo
Utilisez une scène écrite pour le texte vers vidéo ou une image de référence pour l'image vers vidéo. Décrivez une action et le mouvement de caméra. Les contrôles de référence supplémentaires dépendent du modèle sélectionné.
Créez une vidéo FLUX en quatre étapes
Écrivez la scène, choisissez les paramètres pris en charge, vérifiez le coût en crédits et générez. Inspectez votre clip terminé dans l'Historique avant de le télécharger ou de le partager.
1. Écrivez une invite ou téléchargez une image
Commencez par une action, comme un produit qui tourne sous une lumière douce ou un personnage qui traverse une rue. Ajoutez le cadrage, l'éclairage et le mouvement de caméra. Pour l'image-vers-vidéo, utilisez une référence que vous avez le droit d'importer.
2. Choisissez les paramètres et générez
Sélectionnez un modèle vidéo FLUX et utilisez les contrôles proposés pour ce modèle. Vérifiez la durée, la résolution, les options audio et les crédits requis avant de soumettre votre demande.
3. Vérifiez et téléchargez
Ouvrez votre résultat terminé dans l'Historique, lisez le clip et vérifiez le mouvement, les détails et le son. Téléchargez la version que vous souhaitez conserver pour votre workflow de montage ou de publication.
4. Affinez votre prochaine version
Ajustez une partie du prompt ou un paramètre pris en charge, puis comparez le résultat. Conservez les modifications visibles et assemblez les clips individuels dans votre éditeur vidéo préféré.
Idées de vidéo pour votre prochain projet
Transformez un brief créatif ciblé en court matériel visuel pour les contenus sociaux, les récits de produits et les storyboards. Choisissez le format proposé par votre modèle et examinez chaque résultat selon son usage prévu.
Récits de produits et publicité
Montrez un produit dans un cadre spécifique avec une action visible. Utilisez des images de référence lorsque cela est pris en charge, laissez de la place pour la typographie finale et vérifiez les détails du produit avant de publier une publicité.
Clips sociaux et Reels
Construisez un court clip autour d'un seul message pour Shorts, Reels ou TikTok. Choisissez un format vertical ou grand écran pris en charge et gardez le sujet lisible sur un petit écran.
Explications et leçons visuelles
Illustrez une idée à la fois avec une courte scène. Vérifiez les faits et les détails visuels, et ajoutez les légendes ou diagrammes finaux dans votre workflow de montage lorsque l'exactitude est importante.
Présentations créatives
Utilisez un court concept visuel pour communiquer une idée lors d'un pitch ou d'une présentation. Décrivez ce qui doit se produire à l'écran, puis comparez les versions avec votre équipe.
Concepts de film et storyboards
Testez une direction de scène avant de planifier une production plus importante. Donnez à chaque plan un sujet, une action et une direction de caméra, puis vérifiez la continuité lors de l'assemblage d'une séquence.
Anime et récits personnels
Explorez un personnage, une ambiance ou un souvenir sous forme de courte scène animée. Utilisez des directions précises d'éclairage et de style, puis examinez le personnage et les détails dans chaque résultat.
Notes créatives pour votre prochaine vidéo
Donnez à la scène une action principale. Un sujet précis et un mouvement visible rendent un brief vidéo plus facile à évaluer.
Création vidéo FLUX : questions et réponses
Créez de courts clips pour du contenu social, des récits produit et des storyboards créatifs à partir d'un prompt textuel ou d'une image. Vérifiez les entrées et paramètres pris en charge dans le générateur avant de soumettre une requête.
La génération utilise des crédits. Vérifiez votre solde, choisissez un modèle et ses réglages, puis consultez les crédits nécessaires avant de soumettre. Consultez Tarifs pour les options mensuelles, annuelles et les packs de crédits disponibles.
Le menu des modèles affiche les modèles vidéo FLUX disponibles pour la génération sur ce site. Chaque modèle possède ses propres entrées, paramètres et coûts en crédits ; utilisez ce menu pour vérifier vos options actuelles.
Utilisez flux3 video dans un navigateur de bureau ou mobile. Vous pouvez rédiger des prompts, importer des images prises en charge, consulter l'Historique et télécharger les clips terminés sans installer d'application.
Lisez la vidéo et vérifiez le mouvement, les visages, les détails du produit, le texte et le son. Vérifiez les conditions applicables et tout élément involontaire avant d'utiliser le résultat dans un projet.
Le temps de traitement dépend du modèle sélectionné, des paramètres, de la demande et de la charge du service. Suivez l'état de la demande dans l'Historique et revenez au résultat lorsqu'il est terminé.
Oui. Sélectionnez image vers vidéo et téléchargez une image de référence prise en charge que vous avez l'autorisation d'utiliser. Le générateur affiche les entrées acceptées et toute limite de téléchargement. Décrivez le mouvement que vous souhaitez à partir de l'image de départ.
Consultez les conditions applicables de la plateforme et du modèle avant toute utilisation commerciale. Vous êtes responsable des droits sur les images de référence, les personnes identifiables, les logos et tout autre matériel source, ainsi que de la vérification du résultat final avant publication.
Le modèle sélectionné détermine la durée, la résolution, le rapport d'aspect et les contrôles audio affichés dans le générateur. Choisissez parmi ces contrôles et inspectez la vidéo terminée plutôt que de supposer que tous les modèles FLUX prennent en charge les mêmes options.
Les prompts, les références importées et les résultats générés peuvent être stockés pour traiter les demandes et fournir votre Historique. Les cas publiés et les liens médias publics peuvent être consultés et partagés par d'autres personnes. Lisez la Politique de confidentialité avant d'importer du contenu sensible.
Vous pouvez demander un remboursement dans les 10 jours suivant l’achat si vous disposez de crédits inutilisés issus de cet achat et si vous fournissez un motif clair, sous réserve de notre Politique de remboursement. Si vous êtes éligible, un achat inutilisé peut être remboursé intégralement ; un achat partiellement utilisé est remboursé au prorata des crédits inutilisés restants. Les achats entièrement utilisés ainsi que les crédits gratuits ou promotionnels ne sont pas remboursables.
La validité des crédits dépend du pack, de l’abonnement ou de la promotion qui les a émis. Consultez les conditions d’achat et les dates d’expiration dans Paramètres > Crédits. Un renouvellement ou un nouvel achat ne prolonge pas les crédits plus anciens. Utilisez les crédits avant leur date d’expiration indiquée ; ne partez pas du principe que les crédits d’abonnement inutilisés sont reportés indéfiniment.
Les crédits facturés pour une tâche confirmée comme échouée en raison d'une erreur système sont restitués automatiquement. Une tâche encore en cours de traitement n'a pas encore été confirmée comme échouée. Vérifiez son statut et l'historique de vos crédits avant de soumettre à nouveau. Si un échec confirmé affiche toujours un débit non résolu, envoyez un e-mail à support@flux3video.dev avec l'ID de tâche.
Les résultats de l'IA peuvent varier ; un résultat terminé qui diffère de vos attentes n'est donc pas automatiquement considéré comme un échec technique. Les tâches d'image terminées éligibles peuvent être soumises à une évaluation de remboursement en crédits via l'option de remboursement lorsqu'elle est disponible ; une approbation est requise. Pour une vidéo ou un autre problème, contactez support@flux3video.dev avec l'ID de tâche et les détails. Un remboursement en crédits restitue des crédits, et non de l'argent sur votre moyen de paiement.
Envoyez un e-mail à support@flux3video.dev. Pour la facturation, indiquez l'e-mail de votre compte et l'ID de commande ; pour la génération, indiquez l'ID de tâche, la description du problème et les captures d'écran pertinentes. Si vous avez acheté en tant qu'invité, conservez les données du navigateur d'origine et le reçu de paiement afin que nous puissions vous aider à retrouver l'achat.
Créez votre prochaine vidéo avec flux3 video
Partez d'un texte ou d'une image, choisissez un modèle vidéo FLUX et les paramètres pris en charge, et vérifiez le coût en crédits. Inspectez et téléchargez les résultats terminés depuis l'Historique.
flux3video.dev · Mis à jour
Consulter l’historique des mises à jour
- — Ajout de prompts modifiables, de brouillons distincts pour les modes texte et référence et d’une commande pour rétablir l’exemple.
- — Textes et contenu de la page mis à jour.
- — Contenu de la page publié.