跳至主要內容
Kyeo AI
AI 影片模型
目前無法使用

Wan 2.6 Flash

Wan 2.6 Flash 涉及著重快速出片的 Wan 2.6 影片生成家族能力。目前站內生成保持關閉，不會建立任務或扣除點數；頁面保留可核驗背景，並列出可立即使用的替代模型。

目前暫不支援線上使用

此頁面僅供查看模型功能、適用情境與公開資料；該模型不會出現在首頁工作台，也無法提交生成任務。

查看可用模型

示例

Cinematic scene

Style: 8K IMAX. Photorealistic — no CGI/game-engine look. Cinematography: Emmanuel Lubezki × Roger Deakins. Lighting: Natural only, contre-jour backlight, camera on shadow side, atmospheric haze. WB 6500K. Camera: Physical cine lens, 24fps, 180° shutter, handheld only with aggressive operator shake (4–8cm), Dutch angles (12–20°), no stabilization. Color (60:30:10): Bleached asphalt/building canyon 60%, navy/cream transitioning to chrome armor 30%, chrome flashes, dust, visor LED blue 10%. Skin: Pore-level realism, natural imperfections, wet eyes with catchlights. Performance: Hollywood realism with subtle breathing, eye-lines, micro-reactions. Physics: Real gravity, inertia, contact shadows. No floating props. Composition: Rule of thirds + golden ratio. Continuity: Character, props and environment remain identical. No identity drift. Audio: Environmental SFX only. No music. No subtitles. SUBJECT — <<<580cbf5c-4d9d-48b4-bd6d-ca414ac81650>>> matches input 100%, holding <<<d8cb3bc0-32c8-45bd-a5ea-13fa3c59fc7e>>> in right hand. By the end transforms completely into <<<13ce0872-5dbd-4f81-9d78-bb440b750841>>> with armor assembling live in mid-air, matching reference 100%. LOCATION — <<<781dadcd-7026-4651-a9a2-181c3fc22064>>> is STYLE REFERENCE ONLY. Use its mood, architecture and harsh midday lighting for a deserted NYC-style avenue. Extend the world freely with new streets, buildings and angles. Character continuously travels through space. ACTION — 0:00–0:02 — Extreme close-up. Drinks from <<<d8cb3bc0-32c8-45bd-a5ea-13fa3c59fc7e>>>. Visible swallow. Hard cut. 0:02–0:04 — Same framing. Can lowers. Eyes snap open naturally. Genuine grin rapidly forms. Deep inhale. Hard cut. 0:04–0:06 — Wide low angle. Explodes into a 30 km/h sprint straight ahead. Sneakers hit cracked asphalt, dust bursts. Hard cut. 0:06–0:08 — Tracking sprint. Arms pump, jacket trails, grin maintained, accelerating. Hard cut. 0:08–0:10 — Final strides. Right foot cracks asphalt and launches upward at 65 km/h. Hard cut. 0:10–0:13 — Mid-air, full-speed (NO slow motion). Body constantly moving while armor pieces fly from the city and lock onto chest, shoulders, arms, legs and helmet in one continuous transformation. Metallic visor CRACK. Hard cut. 0:13–0:15 — Fully armored <<<13ce0872-5dbd-4f81-9d78-bb440b750841>>> lands in a powerful crouch, dust ring erupts, stands, flexes both gauntlets, glances at chest plate, muffled laugh behind visor. Hard cut to black. CAMERA — Shot 1–2: 18° FOV extreme close, handheld, violent micro-shake. Shot 3–4: 84° FOV, low-angle chase with whip movement and heavy bounce. Shot 5–7: 63° FOV, aggressive handheld tracking, fast reframes, dynamic Dutch angles, impact recoil on landing. CONSTRAINTS — 16:9, MULTISHOT with hard cuts, no slow motion anywhere, no glowing eyes, transformation at full speed, handheld throughout, cut to black at end. <<<781dadcd-7026-4651-a9a2-181c3fc22064>>> is style reference only; do not recreate the input frame. Grin fully established before sprint.

Cinematic scene

生成一段15秒极具戏剧张力的电影级写实剧情短片，展现一个女人在遭遇连续暴力时的绝不屈服，注意情绪递进。 镜头语言与视觉风格：全程保持绝对的固定机位近景特写（LOCKED-ON SHOT），无推拉摇移，将视觉焦点强制锁定在红衣女子的面部。背景为大面积冷灰色调的暗黑地下车库，与女子身上的高饱和度红色蕾丝裙形成极为强烈的视觉冲突，整体色调压抑、紧绷。 动作与叙事节奏： 0-7秒：画面右侧（穿紫色蕾丝袖的手）先后两次扇女子巴掌。每次女子的头因受力被打偏后，都以极其缓慢的节奏转回头，用麻木、隐忍且倔强的眼神冷冷直视施暴者。 7-12秒：暴力节奏骤然加快，施暴者连续高频扇女子约五次耳光，女子头部左右摇摆，发丝散乱，动静对比达到情绪高潮。 12-15秒：动作停止，画面重新归于平静的凝视。女子脸颊微红，碎发贴脸，再次以极度缓慢的速度转头死死盯住对方，眼神透出极度的冰冷与坚不可摧的挑衅。 特殊控制指令：人物面部特征与服装保持绝对一致无跳变，受击时的物理反馈真实自然，微表情清晰细腻。配合画面节奏生成清脆的掌掴音效与极度压抑的底层氛围音，无对白。

Cinematic scene

Overarching Concept: Continuous character journey, moving from gritty Y2K streetwear on a dark rooftop to a stylized, high-tech "tech-ninja" aesthetic inside a pristine white iOS interface dimension. Fast-paced, fluid, and highly interactive with the camera. Universal Rule: The same red-haired woman from Scene 1 must appear in all subsequent scenes, ensuring perfect visual continuity. Scene 1 (00:00–00:01): Rooftop Throw Location: Windy urban rooftop at night, glowing city skyline. Wardrobe (Character 1): Y2K style — baggy vintage denim jacket, graphic baby tee, wide-leg parachute pants. Character Continuity: A specific red-haired woman with wavy copper hair. Action & Camera: Static camera. The red-haired woman spins rapidly, her loose clothes catching the wind. Mid-spin, she throws an iPhone 17 Pro Max directly into the camera lens. Scene 2 (00:01–00:02): Giant Phone Spin Location: Blurred night city backdrop. Action & Camera: Focus pulls to the giant, spinning iPhone 17 Pro Max in the foreground. It stops spinning, hovering vertically and centered. Scene 3 (00:02–00:07): The "Twist & Pull" Transition (CRITICAL FOCUS — NO GLASS BREAKING) Location: Still on the night rooftop, phone centered. Character Continuity: The same red-haired woman (from Scene 1) is visible inside the glowing white phone screen, smiling confidently. Wardrobe (Character 2): A tight, form-fitting, cropped matte-black leather motorcycle jacket over a white top. Action (The Twist): Her hand smoothly emerges from the screen into 3D space (the screen acts like a fluid boundary — no glass breaking). She reaches toward the camera and performs a twisting motion with her wrist and fingers — exactly like screwing in a lightbulb or turning a dial. Camera (Max Zoom): As she twists, the camera zooms in maximally on her hand. Action (The Pull): She reverses the twisting motion and rapidly retracts her hand back into the phone. Camera (The Dive): This reverse motion acts as a physical "pull." As she yanks her hand back, she pulls the camera with her — the camera flies rapidly into the phone screen, fully entering the digital interface. Scene 4 (00:07–00:14): UI Dimension & The Final Punchline Location: Bright, infinite white 3D space. Character Continuity: The same red-haired woman stands center, smiling confidently. Wardrobe (Character 3 — "Tech-Ninja"): Full matte-black tactical jumpsuit in technical fabric; asymmetrical layered vest with utility straps; multi-strap chest harness with glowing magnetic buckles; full gloves with integrated digital displays; chunky combat platform boots. Interface Elements: Gigantic simplified floating 3D app icons — Instagram camera, TikTok "d", Snapchat ghost — swirling around her. The Finale: She looks directly at the camera, playfully catches a glowing "Like" (Heart) icon from the air, gives a sharp wink, and throws the icon straight into the camera lens. The icon covers the lens, turning the screen instantly to black — creating a definitive ending or a seamless loop back to the start.

Cinematic scene

15-second cinematic action sequence aboard a magnetic levitation war train speeding through a dystopian night desert, electric storms illuminating giant industrial ruins in the distance. 0.0–3.0s: Low tracking shot races along the side of the train as a cyber samurai runs across the magnetic exterior panels, sparks erupting beneath his boots. 3.0–6.0s: A chrome-armored assassin descends from a hovering drone transport onto the train roof, neon katana igniting blue light through the storm. 6.0–9.0s: First sword clash sends electrical arcs across the train roof while lightning flashes overhead in slow motion. 9.0–12.0s: High-speed duel across unstable train cars, both fighters nearly thrown off balance as the train tears through debris and collapsing structures. 12.0–15.0s: Massive aerial pull-back reveals the glowing train cutting across the dark wasteland while the duel continues atop the final car.

Cinematic scene

SUBJECTS: Two 2D stick fighters drawn as solid black lines on a pure white field. One is lean and reactive, the other is broader and hits with heavy forward pressure. ENVIRONMENT: Empty white background, no props, only black speed lines, impact circles, smear frames, skid marks, and sharp red blood accents. The negative space keeps every strike readable. MOOD: Brutal, snap-fast, relentless, showy, unfinished. COLOR LOGIC: Monochromatic Grading TIMELINE: 0:00-0:01: Wide shot with extreme negative space. The lean fighter holds center as the broader fighter drops in from above already swinging. Hard push-in. 24mm wide feel. SFX: (air drop, hard land, sharp whoosh). Pure white field, hard black lines only. 0:01-0:03: CUT ON ACTION. Side-on WS. The broader fighter fires punch, hook, knee. The lean fighter slips, blocks, absorbs the knee, then answers with low kick, fast cross, spinning back kick. A thin red blood arc snaps off the last hit. Short truck move. 35mm spherical feel. SFX: (heavy swishes, hit ticks, dry thuds, foot taps). 0:03-0:05: SMASH CUT. Medium wide side view. The broader fighter answers with body hook, elbow, backfist, stomp kick, rushing shoulder, driving the lean fighter across frame. Brief red spatter near the impact rings. Fast lateral move with slight undercrank. 35mm flat animation feel. SFX: (impact cracks, whip swishes, stomp hit, short air bursts). 0:05-0:07: CUT ON ACTION. Side-on WS. The lean fighter ducks under the next swing, steps inside, lands ribs punch, chin elbow, rising chest kick. A brief red spray snaps off the last impact. Short truck move. 35mm film feel. SFX: (duck swish, body thud, elbow crack, kick crack, skid scrape). 0:07-0:10: WHIP PAN TRANSITION. Side-on WS. Both rush back in and trade a twelve-hit string, punch, block, elbow, knee, duck, backfist, kick, catch, shove, hook, spinning elbow, straight cross. Quick red flicks snap off the hardest contacts. Fast truck move with momentum cutting. 35mm spherical feel. SFX: (rapid hit flurry, hard swishes, bass ticks, breath bursts). Thick black impact circles on the heaviest hits. 0:10-0:12: CUT ON ACTION. Worm's eye wide shot. The broader fighter surges again, the lean fighter absorbs two hits, slips the third, steps off the planted thigh, vaults up, hammers down, and both collide midair in a tight X silhouette that cracks apart instantly. Arc shot around the clash. 24mm wide feel. SFX: (rising whoosh, hit flurry, double crack, burst hiss, bass slam). 0:12-0:15: HARD CUT. Monumental WS. Both land already moving, skid one step, rebound, and rush back to center without resetting, disappearing into black strike smears and red flecks over the last impacts. Tiny push-in, ending mid-exchange. 50mm standard feel. SFX: (skid scrape, hard swishes, dry impacts, low ring).

Cinematic scene

Slow-motion extreme hyper close-up, 16:9. Shaky organic handheld shot pushed in tight and intimate on the face of a man in his thirties with dark wet side-parted hair plastered to his forehead, mid-lunge through the flood his face fills the entire frame, jaw clenched with teeth bared and gritted in a snarl of effort, brow deeply furrowed, eyes narrowed and locked forward with raw desperate determination. Water streams and beads down his skin, droplets clinging to his eyelashes and running off his chin, spray bursting across his face in slow suspended flecks. Every pore, water bead, and strand of wet hair rendered in sharp tactile detail. The chaotic flooded office and blurred panicking crowd fall completely out of focus behind him. In slow motion the droplets hang and drift, sweat and water glistening. Dynamic unsteady handheld movement right at his face, breathing with the motion. Sickly institutional palette muted seafoam green and cold reflective water bokeh behind him. Face anatomically correct, stable, hyperrealistic, genuine intense emotion, natural human proportions, no warping or distortion. Heavy coarse analog 35mm film grain baked into every frame, thick and consistent, grain crawling over highlights and shadows, like a scanned celluloid film print, never clean, never digital-looking. Flat overhead fluorescent light, dramatic slow-motion intensity, shallow depth of field, cinematic photorealistic, ultra-detailed.

這個模型表現如何？

這是一頁狀態清楚的模型資料，而不是生成入口。你可以先了解它解決什麼問題、為何目前不可用，以及如何把相同需求移轉到已啟用模型。

它擅長什麼
Wan 2.6 Flash 的公開定位是著重快速出片的 Wan 2.6 影片生成家族能力。這項家族能力不代表站內目前已具備可呼叫合約。
適合哪些使用者
適合正在研究該模型、核對目前狀態，或需要尋找同類可用方案的使用者。
為什麼要在 Kyeo AI 使用
頁面保持可索引，是為了提供準確狀態與替代路徑；生成、報價和附件入口繼續關閉。
清楚標示目前不可用不展示未核驗價格或參數提供已啟用替代模型
Wan 2.6 Flash 目前狀態Wan 2.6 Flash 是否可用Wan 2.6 Flash 替代模型Wan 2.6 Flash 能力說明

使用前須知

目前無法從此頁發起生成，也不會建立任務或扣除點數。
歷史介面、價格或參數不代表今天的站內合約，因此不會當作可用承諾展示。
重新開放前必須再次核驗呼叫、上傳、計費、失敗退款和多語內容。

同類模型推薦

還沒決定該用哪個嗎？順便比較一下這幾個同類候選模型。

Wan 2.6

不附媒體時使用文字生成影片，恰好 1 張圖片時使用圖片生成影片，1–3 個影片時使用影片編輯；圖片與影片不可混用。 預設為 1080p、5 秒、鏡頭結構關閉、內容過濾開啟。

查看詳情

Wan 2.7 Video

Wan 2.7 是一款涵蓋文生影片、首幀或首尾幀圖生影片和影片編輯的 AI 影片模型。目前可選 720p 與 1080p；選型時應一併比較素材要求、時長、畫幅和預估點數。

查看詳情

Wan 2.2 A14B Turbo

Wan 2.2 A14B Turbo 在 Kyeo 並非單一工作流程：沒有素材時使用文字生成，上傳一張圖片後改走圖片生成，同時提供一張圖片與一段音訊才會進入語音驅動。三條工作流程的解析度、點數與可見參數不同，選擇前應先確認素材組合。

查看詳情

常見問題

資訊來源

本頁區分模型家族的公開能力與站內目前可用性。未核驗的價格、參數和接入狀態不會寫成現行承諾。

站內可用性複核
串接

最近一次複核確認該模型仍不可呼叫；頁面只保留狀態說明、能力背景和已啟用替代項。