已驗證提示詞
a] Type: Creative Brief. Genre: Hyper-realistic Surrealism. Composition Style: Composite Portrait. Aspect Ratio: Portrait (implied by 'portrait' description). [Scene Architecture] Viewpoint: [Type] Photographic. [Angle] High-angle / Looking down. [Framing] Tight on central subject.. Dimensional Hierarchy: [Rule] Scale disparity for surreal effect. [Dominant Element] iPhone 17 Pro Max (Super-scaled). [Subordinate Elements] Blue Book (Miniature), Pen (Miniature).. [Realm Physical] Description: The real-world environment surrounding the device.. Environment: [Surface] Wooden table. [Texture Attributes] rich grain, tactile, worn.. Lighting Global: [Source] Natural light. [Temperature] Warm. [Shadow Quality] Soft, diffused, volumetric.. Active Agent: [Identity] Human Hand (Real). [Action] Pouring. [Position] Entering frame laterally.. Held Object: [Item] Bottle. [State] Chilled (visible condensation). [Branding] Logo Text: Decamin. Placement: Visible on label. [Contents] Substance: Water. Color: Light Green. State: Liquid flow.. Static Props: [Item] Book. [Color] Blue. [Scale Notes] Significantly smaller than phone., [Item] Pen. [Type] Ballpoint/Ink. [Scale Notes] Significantly smaller than phone.. [Realm Digital] Description: The content displayed on the screen.. Container Device: [Model] iPhone 17 Pro Max. [State] Screen ON. [Orientation] Flat on physical surface.. Screen Content: [Subject Identity] mqn (Reference ID). [Subject Scale] Close-up (filling screen). [Expression] Happy / Smiling. [Attire] Winter clothing (matching reference). [Setting] Winter landscape / snowy backdrop. [Held Object Digital] Item: Drinking Glass. Branding: [Logo Text] Decamin. [Visibility] Clear.. Initial State: Empty (waiting for pour).. [Surreal Bridge Event] Description: The interaction connecting the physical and digital realms.. Action Type: Trans-dimensional Fluid Dynamics. Source: realm_physical.held_object.contents (Light Green Water). Interaction Point: realm_digital.container_device.screen_surface. Destination: realm_digital.screen_content.held_object_digital (The Glass). Physics Violation Rules: [Rule 1] Liquid does not splash off the glass screen surface.. [Rule 2] Screen surface acts as a permeable membrane solely for this liquid.. [Rule 3] Physical liquid transitions seamlessly into digital representation upon contact... Visual Details: Sharp liquid simulation, No surface tension on screen glass, Fluid physically filling digital cup. [Rendering Specifications] Visual Fidelity: Hyper-realistic. Texture Focus: Sharp fluid details, Glass pixels, Wood grain, Skin texture (hand and subject). Mood: Cinematic, warm, magical. Resolution Target: 8K / Highly detailed.
a] Type: Creative Brief. Genre: Hyper-realistic Surrealism. Composition Style: Composite Portrait. Aspect Ratio: Portrait (implied by 'portrait' description). [Scene Architecture] Viewpoint: [Type] Photographic. [Angle] High-angle / Looking down. [Framing] Tight on central subject.. Dimensional Hierarchy: [Rule] Scale disparity for surreal effect. [Dominant Element] iPhone 17 Pro Max (Super-scaled). [Subordinate Elements] Blue Book (Miniature), Pen (Miniature).. [Realm Physical] Description: The real-world environment surrounding the device.. Environment: [Surface] Wooden table. [Texture Attributes] rich grain, tactile, worn.. Lighting Global: [Source] Natural light. [Temperature] Warm. [Shadow Quality] Soft, diffused, volumetric.. Active Agent: [Identity] Human Hand (Real). [Action] Pouring. [Position] Entering frame laterally.. Held Object: [Item] Bottle. [State] Chilled (visible condensation). [Branding] Logo Text: Decamin. Placement: Visible on label. [Contents] Substance: Water. Color: Light Green. State: Liquid flow.. Static Props: [Item] Book. [Color] Blue. [Scale Notes] Significantly smaller than phone., [Item] Pen. [Type] Ballpoint/Ink. [Scale Notes] Significantly smaller than phone.. [Realm Digital] Description: The content displayed on the screen.. Container Device: [Model] iPhone 17 Pro Max. [State] Screen ON. [Orientation] Flat on physical surface.. Screen Content: [Subject Identity] mqn (Reference ID). [Subject Scale] Close-up (filling screen). [Expression] Happy / Smiling. [Attire] Winter clothing (matching reference). [Setting] Winter landscape / snowy backdrop. [Held Object Digital] Item: Drinking Glass. Branding: [Logo Text] Decamin. [Visibility] Clear.. Initial State: Empty (waiting for pour).. [Surreal Bridge Event] Description: The interaction connecting the physical and digital realms.. Action Type: Trans-dimensional Fluid Dynamics. Source: realm_physical.held_object.contents (Light Green Water). Interaction Point: realm_digital.container_device.screen_surface. Destination: realm_digital.screen_content.held_object_digital (The Glass). Physics Violation Rules: [Rule 1] Liquid does not splash off the glass screen surface.. [Rule 2] Screen surface acts as a permeable membrane solely for this liquid.. [Rule 3] Physical liquid transitions seamlessly into digital representation upon contact... Visual Details: Sharp liquid simulation, No surface tension on screen glass, Fluid physically filling digital cup. [Rendering Specifications] Visual Fidelity: Hyper-realistic. Texture Focus: Sharp fluid details, Glass pixels, Wood grain, Skin texture (hand and subject). Mood: Cinematic, warm, magical. Resolution Target: 8K / Highly detailed.
已驗證提示詞
Borrowed from @iX00AI: Using the uploaded reference image, generate a horizontal three-panel composition with mini-skirt Santa cosplay in winter snow theme.
Borrowed from @iX00AI: Using the uploaded reference image, generate a horizontal three-panel composition with mini-skirt Santa cosplay in winter snow theme.
已驗證提示詞
A flat branded paper folds itself into the full 3D shape of a [insert product or item, e.g. “Coca-Cola bottle”, “Nike sneaker”, “Big Mac”], mid-motion. The paper colors match the [insert brand name] brand’s signature palette and the natural colors of the item (e.g., [describe key colors or ingredients, like “red and white for Coca-Cola”, “brown, green, yellow for Big Mac”]). Dramatic studio lighting, origami texture detail, soft gradient shadows. Stylized with Japanese minimalism and elegant negative space. The scene captures a clean, elevated transformation from flat brand identity into sculptural product form.
A flat branded paper folds itself into the full 3D shape of a [insert product or item, e.g. “Coca-Cola bottle”, “Nike sneaker”, “Big Mac”], mid-motion. The paper colors match the [insert brand name] brand’s signature palette and the natural colors of the item (e.g., [describe key colors or ingredients, like “red and white for Coca-Cola”, “brown, green, yellow for Big Mac”]). Dramatic studio lighting, origami texture detail, soft gradient shadows. Stylized with Japanese minimalism and elegant negative space. The scene captures a clean, elevated transformation from flat brand identity into sculptural product form.
已驗證提示詞
已驗證提示詞
Colossal marble statue of [INSERT OBJECT OR CHARACTER HERE], wings (if applicable) spread in divine majesty, adorned with ornate golden armor and flowing baroque robes, cracked marble textures with soft patina, glowing sacred emblem on chest, standing within a celestial golden dome, surrounded by divine architecture, dramatic god rays from above, dynamic mid-action pose, ultra-detailed renaissance sculpture style, cinematic lighting, majestic and sacred atmosphere, 8k resolution
Colossal marble statue of [INSERT OBJECT OR CHARACTER HERE], wings (if applicable) spread in divine majesty, adorned with ornate golden armor and flowing baroque robes, cracked marble textures with soft patina, glowing sacred emblem on chest, standing within a celestial golden dome, surrounded by divine architecture, dramatic god rays from above, dynamic mid-action pose, ultra-detailed renaissance sculpture style, cinematic lighting, majestic and sacred atmosphere, 8k resolution
已驗證提示詞
A cinematic, intimate portrait, taken in front of a large, bright window. A beautiful East Asian woman, with the visual appeal of a K-pop girl group member, 175cm tall and weighing 50kg, possesses a perfectly proportioned figure. She leans against the window frame in an innocent profile, glancing back at the camera with a gentle, shy smile. She wears a white cotton low-cut tank top. Strong sunlight streams down from behind her, penetrating her loosely styled hair and creating a dazzling halo around her shoulders. The image is sharply focused, clearly showing the fine downy hair on her cheeks and nose under the backlight.
A cinematic, intimate portrait, taken in front of a large, bright window. A beautiful East Asian woman, with the visual appeal of a K-pop girl group member, 175cm tall and weighing 50kg, possesses a perfectly proportioned figure. She leans against the window frame in an innocent profile, glancing back at the camera with a gentle, shy smile. She wears a white cotton low-cut tank top. Strong sunlight streams down from behind her, penetrating her loosely styled hair and creating a dazzling halo around her shoulders. The image is sharply focused, clearly showing the fine downy hair on her cheeks and nose under the backlight.
已驗證提示詞
已驗證提示詞
已驗證提示詞
Create the highest-resolution picture of me posing elegantly with my back slightly turned and my head looking over the shoulder towards the camera. I am wearing a sleek, form-fitting strapless black evening gown, paired with long, elegant black velvet opera gloves. My hair is styled in loose, glossy waves cascading naturally over my shoulder. I wear small, subtle earrings and a delicate bracelet on one wrist. My expression is confident, graceful, and alluring. The background is a plain stucco gray with soft gradient lighting, giving it a high-fashion photography look. The lighting is dramatic but soft, highlighting the contours of my body and face with realistic shadows and textures. The image should look like a high-quality, ultra-realistic fashion magazine photo, keeping my real face exactly as shown in the reference photo.
Create the highest-resolution picture of me posing elegantly with my back slightly turned and my head looking over the shoulder towards the camera. I am wearing a sleek, form-fitting strapless black evening gown, paired with long, elegant black velvet opera gloves. My hair is styled in loose, glossy waves cascading naturally over my shoulder. I wear small, subtle earrings and a delicate bracelet on one wrist. My expression is confident, graceful, and alluring. The background is a plain stucco gray with soft gradient lighting, giving it a high-fashion photography look. The lighting is dramatic but soft, highlighting the contours of my body and face with realistic shadows and textures. The image should look like a high-quality, ultra-realistic fashion magazine photo, keeping my real face exactly as shown in the reference photo.
已驗證提示詞
A vibrant, hyper-realistic miniature of [CITY] inside a transparent capsule lying on a sandy beach. The capsule is half [COLOR] (matching [COUNTRY]’s flag) with [CITY] written in white text on the colored section. Inside the capsule: iconic landmarks of [CITY], beautiful water canals or streets, small detailed boats or cars, sunny bright lighting, cinematic depth of field, dreamy atmosphere, ocean waves in the background.
A vibrant, hyper-realistic miniature of [CITY] inside a transparent capsule lying on a sandy beach. The capsule is half [COLOR] (matching [COUNTRY]’s flag) with [CITY] written in white text on the colored section. Inside the capsule: iconic landmarks of [CITY], beautiful water canals or streets, small detailed boats or cars, sunny bright lighting, cinematic depth of field, dreamy atmosphere, ocean waves in the background.
已驗證提示詞
a-detailed render of a [CHARACTER NAME] paper toy version in box form (papertoy) made from matte folded cardboard and cut with visible paper texture, clean edges and neat folds. Cubic head and body, square extremities, simplified facial features, flat printed colors and subtle shading for greater depth. The clothing and accessories faithfully imitate the appearance of the reference image in a minimalist geometric papercraft style, maintaining compact proportions and chibi style. Neutral studio lighting, soft shadows, smooth background, photorealistic product photography, 4K, no text or logos. 1080x1080 dimension.
a-detailed render of a [CHARACTER NAME] paper toy version in box form (papertoy) made from matte folded cardboard and cut with visible paper texture, clean edges and neat folds. Cubic head and body, square extremities, simplified facial features, flat printed colors and subtle shading for greater depth. The clothing and accessories faithfully imitate the appearance of the reference image in a minimalist geometric papercraft style, maintaining compact proportions and chibi style. Neutral studio lighting, soft shadows, smooth background, photorealistic product photography, 4K, no text or logos. 1080x1080 dimension.
已驗證提示詞
rreal yet hyper-realistic image in levitation photography style, where elements of [subject] float chaotically and randomly above an outstretched hand, interwoven with symbolic elements that directly oppose it. The background forms dynamically to visually embody the symbolism of the subject and its opposite, through blurred or abstract details that reflect either tension or harmony between the two—like a mirror inside a dream. Each element floats at a unique angle and varying distance. Background: Shifts according to the subject and its opposite Lighting: Soft cinematic with dreamy shadows Style: Hyper-realistic + Levitation Photography + Fantasy Aspect Ratio: 1:1
rreal yet hyper-realistic image in levitation photography style, where elements of [subject] float chaotically and randomly above an outstretched hand, interwoven with symbolic elements that directly oppose it. The background forms dynamically to visually embody the symbolism of the subject and its opposite, through blurred or abstract details that reflect either tension or harmony between the two—like a mirror inside a dream. Each element floats at a unique angle and varying distance. Background: Shifts according to the subject and its opposite Lighting: Soft cinematic with dreamy shadows Style: Hyper-realistic + Levitation Photography + Fantasy Aspect Ratio: 1:1
為你的下一個影片起始畫格而設的 FLUX AI 圖像生成器
以清晰的視覺方向建立靜態圖像
在 flux3 影片上使用 FLUX 圖像模型來探索主體、產品場景或分鏡畫格。從文字提示詞或支援的圖像參考開始,檢視設定和點數費用,並在將其用於影片之前檢查結果。
建立你的第一張圖像在規劃動態之前先建立靜態圖像
FLUX AI 圖像生成器可協助你為影片開發視覺方向。這些是圖像模型:在此建立或精進你的靜態圖像,然後使用獨立的影片生成器將合適的起始畫格製作成動畫。
描述你的主體與構圖
說明主體、其位置、場景、光線和預期風格。加入重要的顏色或材質,然後選擇所選圖像模型可用的輸出控制項。
發展一致的視覺風格
描述你為分鏡或行銷活動想要的調色、渲染風格和氛圍。根據相同的簡介比較圖像,並精進具體細節,而不是假設每個結果都會相符。
精進支援的圖像參考
當你所選的 FLUX 模型支援圖像編輯時,上傳你有權使用的參考。說明一項變更和要保留的細節。可用的參考輸入取決於模型。
為圖像轉影片準備起始畫格
建立一張為預期動作和攝影機運動留出空間的靜態圖像。下載你想使用的結果,開啟影片生成器,並在圖像轉影片模式中上傳。影片會在該獨立工作流程中生成。
在分享前檢視細節
檢查臉部、產品形狀、文字和任何非預期的物件。檢視適用的模型和平台條款,並在發布或在專案中使用圖像前確認你對參考的權利。
將你的創意簡介保留在一個工作區中
撰寫提示詞,選擇 FLUX 圖像模型,檢視其點數費用,並在歷史記錄中找到完成的作品。保留你想用作影片起點的版本。
塑造 FLUX 圖像簡介的五種方法

影片創作者的起始影格構想
探索場景、產品故事、社群短片或插畫概念的構圖與燈光研究。已發布的影像參考僅用於說明視覺方向,而非保證的轉換結果。


場景與分鏡影格
描述環境、主體位置、鏡頭大小與光線。使用影像來討論場景方向,並規劃你希望在後續影片中呈現的動作。

產品故事
圍繞單一產品與明確的視覺構想建立場景。在將結果用於廣告之前,請對照實際產品檢查形狀、顏色、材質與任何品牌標示。

社群內容
為貼文或短片開頭建立靜態影像。選擇主體清晰可辨的支援構圖,並在你偏好的剪輯流程中加入最終文字排版。
動畫概念
為動畫場景探索角色、設定與色調。指定你想保留的細節,並在發展成序列之前檢視每個結果。
四個步驟建立影片起始影格
撰寫你的簡述或加入參考
描述主體、構圖、燈光與視覺風格。若所選模型支援編輯,請上傳你擁有使用權限的參考並說明你需要的變更。
選擇 FLUX 影像模型與設定
從可用的影像目錄中選擇模型,並使用其支援的尺寸、品質與參考控制項。提交前請先查看所需的點數。
生成並檢視靜態影像
選擇「生成影像」並在歷史記錄中查看請求。檢查結果,若需要另一個版本,請修改提示或支援的設定。
下載並在影像轉影片中使用
下載你選擇的靜態影像,開啟影片生成器,並將其上傳為起始影格參考。另外描述動作;影像建立與影片建立使用各自的模型與點數報價。
你下一個專案的視覺構想
在規劃社群短片、產品故事、解說或動畫場景之前,先使用靜態影像確立主體、風格與構圖。
角色與風格研究
發展場景的角色、姿勢、燈光與背景。在規劃角色動作之前,先檢視臉部、手部、服裝與細節。
產品與活動概念
從聚焦的簡述探索產品場景或活動氛圍。對照實際產品細節檢查結果,並為最終文字或品牌標示保留空間。
解說與簡報
為一個教學或簡報重點準備簡單的視覺。檢視事實與細節,並在準確性重要時自行加入最終標籤。
環境與故事世界
建立具有明確色調、燈光與尺度感的環境概念。用它來規劃場景方向,再生成個別的影片鏡頭。
起始影格的創意筆記
在加入風格詞之前,先命名主體與場景。讓主要視覺構想保持易於辨識。
FLUX AI 影像生成器:問題與解答
它使用 FLUX 影像模型,從文字提示或支援的影像參考建立靜態影像。在 flux3 video 上,用它來發展視覺與起始影格。影片生成則在影片生成器中另行進行。
描述主體、場景、構圖、燈光與預期風格。當顏色或材質重要時,請一併包含。若為影片起始影格,也請考量主體進行下一個動作所需的空間。
影像模型選單會顯示本網站目前可用的 FLUX 模型。每個模型都有各自的輸入需求、設定與點數成本。生成前請使用該選單查看你的選項。
當所選模型提供時,請使用影像轉影像或編輯模式。上傳你擁有使用權限的支援參考、描述一項變更,並指出要保留的細節。輸入計數與編輯控制項取決於所選模型。
可以。下載你選擇的影像,並在獨立影片生成器的影像轉影片模式中上傳。在那裡描述預期的動作與鏡頭移動。FLUX 影像模型產生靜態影像;影片模型產生片段。
檢視適用的平台與模型條款,以及你對上傳參考、可辨識人物、標誌與其他來源素材的權利。在將影像用於商業專案之前,請檢查是否有非預期的細節。
生成會使用點數。查看你的餘額、選擇 FLUX 影像模型與設定,並在提交前檢視報價費用。前往定價頁面查看可用的方案與點數包。
完成的影像可在歷史記錄中取得。開啟結果以檢視,並下載你想保留用於專案或影像轉影片流程的版本。
若您該筆購買尚有未使用的點數,並提供明確理由,您可以在購買後 10 天內申請退款,惟須符合我們的退款政策。若符合資格,未使用的購買可全額退款;部分使用的購買則依剩餘未使用點數比例退款。已完全使用的購買,以及免費或促銷點數,均不予退款。
點數的有效期限取決於發行該點數的點數包、訂閱或促銷活動。請在「設定 > 點數」查看購買條款與到期日。續訂或新購買不會延長老點數的效期。請在標示的到期日前使用點數;請勿假設未使用的訂閱點數會無限期遞延。
若任務經確認因系統錯誤而失敗,已扣除的點數會自動退回。仍在處理中的任務尚未確認失敗。再次提交前,請查看其狀態與您的點數紀錄。若確認失敗後仍顯示未解決的扣款,請寄電子郵件至 support@flux3video.dev 並提供任務 ID。
AI 生成結果可能有所差異,因此已完成的結果若與您的預期不同,不會自動視為技術故障。符合條件的已完成圖片任務,可使用可用的退款選項提交點數退款審核;須經核准。若為影片或其他問題,請聯絡 support@flux3video.dev 並提供任務 ID 與詳細資訊。點數退款退還的是點數,而非退回您付款方式的款項。
請寄電子郵件至 support@flux3video.dev。若是帳務問題,請提供您的帳戶電子郵件與訂單 ID;若是生成問題,請提供任務 ID、問題描述及相關截圖。若您是以訪客身分購買,請保留原始瀏覽器資料與付款收據,以便我們協助找出該筆購買。

flux3video.dev · 更新日期
查看更新記錄
- — 新增可編輯的範例提示詞,文字與參考模式各自保留草稿,並可還原範例。
- — 已更新頁面文案與內容。
- — 頁面內容已發布。














