Kyeo 的 Veo3.1 Fast 等同 Google 官方 Fast 嗎？

Kyeo 的 Veo3.1 Fast 等同 Google 官方 Fast 嗎？ — 本頁描述目前已驗證的站內合約，不公開也不依賴內部供應商模型識別碼。

`REFERENCE_2_VIDEO` 目前支援幾張圖？

`REFERENCE_2_VIDEO` 目前支援幾張圖？ — 參考圖片模式接受 1～3 張經簽章驗證的 JPEG 或 PNG，只支援 16:9／9:16，且時長必須為 8 秒。

1080p 和 4K 為什麼不是基礎參數？

1080p 和 4K 為什麼不是基礎參數？ — 升解析度操作只接受本人成功的 720p 基礎任務或 Extend 結果；直接生成的 1080p、4K 結果不能再次要求升解析度。

目前可以直接使用純文字生成嗎？

目前可以直接使用純文字生成嗎？ — 純文字、首尾影格與 1～3 張參考圖片路徑會依已提交圖片選擇對應生成模式。

參考圖片能保證角色或風格一致嗎？

參考圖片能保證角色或風格一致嗎？ — 音訊能力仍屬實驗性且不保證可用；交付前必須逐項人工檢查結果。

AI 影片模型

Google

Veo

Veo3.1 Fast

Q: 基礎生成與 Extend 各消耗多少點數？

基礎生成與 Extend 各消耗多少點數？ — 直接生成 720p 扣 30 點數，1080p 扣 38 點數（37.5 無條件進位），4K 扣 90 點數。Extend 扣 30；符合條件的 1080p、4K 結果操作另扣 8、60 點數。

Kyeo AI 的 Veo3.1 Fast 支援純文字、首尾影格與參考圖片生成，可直接選擇 720p、1080p、4K，以及 4、6、8 秒時長。依解析度為 30 / 38 / 90 點數.

使用此模型查看點數費用回到模型廣場

能力標籤

TEXT / FRAME / REFERENCE

點數消耗

720p 30 · 1080p 38 · 4K 90

提示詞上限

5,000 字元

上傳限制

JPEG / PNG · 1–3 · 10 MB

30 秒快速瞭解

純文字、首尾影格與 1～3 張參考圖片路徑會依已提交圖片選擇對應生成模式。 TEXT_2_VIDEO / FIRST_AND_LAST_FRAMES_2_VIDEO / REFERENCE_2_VIDEO.
直接生成 720p 扣 30 點數，1080p 扣 38 點數（37.5 無條件進位），4K 扣 90 點數。Extend 扣 30；符合條件的 1080p、4K 結果操作另扣 8、60 點數。 720p / 1080p / 4K.
參考圖片模式接受 1～3 張經簽章驗證的 JPEG 或 PNG，只支援 16:9／9:16，且時長必須為 8 秒。 REFERENCE_2_VIDEO.
音訊能力仍屬實驗性且不保證可用；交付前必須逐項人工檢查結果。 4 / 6 / 8 s.

熱門搜尋: Veo3.1 Fast 怎麼用 · Veo3.1 Fast REFERENCE_2_VIDEO 是什麼 · Veo3.1 Fast 價格

示例

Cinematic scene

Create a professional cinematic jewelry advertisement storyboard board. Aspect ratio: 16:9 horizontal layout. Theme: Pearl and crystal jewelry elegance. Duration: 15 seconds. Style: High-end fashion editorial presentation, cinematic jewelry film, soft feminine atmosphere, minimal modern studio aesthetic, refined craftsmanship, elegant beauty photography. Color palette: Pearl white, soft champagne gold, warm ivory, black velvet, soft blush pink, subtle silver highlights. Visual direction: Delicate pearl textures, sparkling gemstone reflections, silky fabric movement, soft candle-like highlights, clean elegant compositions, beautiful macro details. No brand. No logo. No text. No watermark. No readable symbols. --- MAIN SUBJECT: A young adult Asian woman with an elegant and natural appearance. Appearance: Soft refined facial features. Long dark hair with gentle waves. Natural glowing skin. Calm confident expression. Graceful posture. Minimal elegant makeup. Wardrobe: Flowing ivory silk dress. Soft fabric texture. Elegant modern fashion silhouette. No visible patterns or logos. --- JEWELRY DESIGN: A delicate pearl and gemstone jewelry collection. Includes: Pearl necklace. Pearl earrings. Elegant gemstone bracelet. Fine metallic details. Design: Smooth natural pearl surfaces. Clear crystal reflections. Fine metal craftsmanship. Minimal timeless shapes. No recognizable brand designs. --- STORYBOARD STRUCTURE: 8 SHOTS · 15 SECONDS · CINEMATIC JEWELRY FILM --- SHOT 1 — THE INTRODUCTION Duration: 0:00–0:02 Wide cinematic shot. The woman stands beside a large window in a softly illuminated room. Morning light enters through sheer curtains. A jewelry display rests nearby on a dark velvet surface. Camera: Slow elegant push-in. Mood: Quiet elegance. --- SHOT 2 — THE PEARL DETAIL Duration: 0:02–0:04 Extreme macro shot. A single pearl necklace rests on soft fabric. Light moves slowly across the pearl surface. Show: smooth texture, soft reflections, delicate craftsmanship. Camera: Macro beauty photography. --- SHOT 3 — THE HANDCRAFT MOMENT Duration: 0:04–0:06 Close-up. The woman's hands gently pick up pearl earrings. Her fingers move carefully. The gemstones reflect soft light. Camera: Slow focus pull. Lighting: Warm highlights. --- SHOT 4 — THE PREPARATION Duration: 0:06–0:08 Medium beauty shot. The woman puts on the pearl earrings. Her hair moves naturally. The jewelry catches the light. Camera: Elegant side profile. Atmosphere: Soft and graceful. --- SHOT 5 — THE REFLECTION Duration: 0:08–0:10 Close-up portrait. The woman looks into a mirror. Pearls and gemstones create subtle reflections. She smiles gently. Camera: Slow cinematic orbit. --- SHOT 6 — THE JEWELRY MOVEMENT Duration: 0:10–0:12 Macro cinematic shot. The necklace moves slightly as she walks. Pearls catch the surrounding light. Fabric flows naturally. Camera: Slow motion fashion movement. --- SHOT 7 — THE COMPLETE LOOK Duration: 0:12–0:14 Full-body fashion composition. The woman walks through an elegant bright interior space. Soft curtains move behind her. Jewelry remains the visual focus. Camera: Smooth tracking shot. --- SHOT 8 — FINAL HERO FRAME Duration: 0:14–0:15 Final product beauty shot. Pearl necklace and earrings displayed on velvet surface. The woman softly appears in the background. Warm light creates a timeless atmosphere. Camera: Slow push-in. --- LIGHTING: Soft diffused lighting. Pearl-like highlights. Warm champagne reflections. Gentle shadows. Cinematic depth. Elegant glow. --- CAMERA STYLE: Fashion editorial camera. Macro jewelry photography. Slow dolly movement. Smooth tracking. Shallow depth of field. Professional commercial composition. --- NEGATIVE: brand name, logo, watermark, text, letters, celebrity, famous person, recognizable jewelry design, copyrighted design, exaggerated luxury symbols, money, wealth symbols, unrealistic gemstones, plastic skin, distorted hands, bad anatomy, low quality, blurry.

Cinematic scene

Create a 30-second, 1080p ultra-realistic home video of the same young adult woman spending an unplanned afternoon in a quiet coastal neighborhood. Character: Natural expressive face, warm brown eyes, realistic skin, long dark-brown messy ponytail. Ivory oversized linen shirt, muted-blue camisole, sand-colored trousers, white sneakers, brown crossbody bag, silver hoops and watch. Maintain exact identity, outfit, hairstyle, proportions, and accessories throughout. Setting: Authentic seaside residential streets with weathered homes, narrow lanes, plants, bicycles, laundry, ocean glimpses, and a waterfront path. No brands, landmarks, advertisements, or staged crowds. Camera: Older handheld consumer camera operated by a friend. Imperfect framing, natural shake, autofocus hunting, exposure shifts, motion blur, digital noise, and compression. No gimbal, drone, slow motion, or polished commercial look. Sequence: 0–5s: She exits carrying a paper bag, checks her phone, and laughs when wind catches her shirt. 5–10s: She notices colorful kites, stops to watch, and the camera briefly loses focus. 10–15s: A child's kite catches on a fence; she gently frees it and returns it, creating a natural spontaneous moment. 15–20s: She walks toward the sea and quietly watches the water as wind moves her hair and clothing. 20–25s: Ocean spray surprises her; she laughs, looks toward the camera, and wipes water from her face. 25–30s: Warm sunlight breaks through clouds. She smiles subtly at the camera, then walks away beside the ocean. Hard cut to black at 29.5s. Realism: Accurate physics, natural walking, hair, fabric, paper, kite strings, seawater, lighting, shadows, hands, and facial expressions. No distortions, duplicated objects, disappearing props, wardrobe changes, or anatomy errors. Audio: Natural wind, ocean, birds, footsteps, bicycles, neighborhood ambience, paper rustling, and genuine laughter. No music, narration, subtitles, captions, logos, or watermarks. Final feel: A beautiful but ordinary moment accidentally captured by a friend—intimate, imperfect, spontaneous, and genuinely real.

Cinematic scene

SEEDANCE 2.5 | 30s | 16:9 | ULTRA-PHOTOREALISTIC SMARTPHONE VLOG A genuine personal smartphone travel vlog of a Korean woman in her early 20s experiencing Istanbul for the first time. Spontaneous, natural and slightly imperfect — never a tourism commercial. CHARACTER: Same Korean woman throughout. Petite oval face, expressive brown eyes, long black hair with loose strands, realistic fair skin and natural imperfections. Cream ribbed top, dark-brown wide-leg trousers, open beige trench coat, brown crossbody bag, white sneakers, delicate gold jewelry. No identity or wardrobe changes. She speaks only Turkish with a natural Korean accent, slightly imperfect and conversational. No subtitles, captions or translations. CAMERA: Handheld smartphone, 16:9 landscape. Natural arm movement, walking jolts, imperfect framing, autofocus delays, exposure shifts, realistic motion blur and phone-camera texture. No gimbal, drone, cinematic lens effects or commercial grading. STORY 0–4s — HAGIA SOPHIA: Selfie walk through Sultanahmet with Hagia Sophia behind her. She turns the camera toward it, then back to herself and says: “Vay… Ayasofya gerçekten çok güzel.” 4–7s — BLUE MOSQUE: Show the Blue Mosque, square, pigeons and pedestrians. She walks through frame, looking around with quiet curiosity. 7–11s — TOPKAPI: She explores Topkapı Palace courtyards and architecture, briefly showing the Bosphorus. She says: “Burası çok eski… çok güzel.” 11–14s — BREAKFAST: Local café table with simit, cheese, olives, eggs and Turkish tea. She takes a bite of simit and says: “Bu kahvaltı çok güzel!” 14–17s — COFFEE: Close handheld shot of Turkish coffee being prepared and poured. She sips, reacts naturally and says: “Kahve çok güçlü!” She laughs. 17–20s — GRAND BAZAAR: Busy authentic bazaar with lamps, ceramics, textiles and pedestrians. She films while walking and casually moves aside as someone passes. 20–23s — GALATA / İSTİKLAL: Galata Tower, historic streets and red tram. She looks up, turns to camera and says: “İstanbul çok güzel bir şehir.” 23–27s — BOSPHORUS: Golden-hour ferry selfie. Wind moves her hair and trench coat. She turns toward the Bosphorus, ferries, seagulls and bridges, then softly says: “Burası inanılmaz…” 27–30s — NIGHT: Warm Istanbul street lights, taxis, pedestrians and trams behind her. She smiles at the camera and says: “İstanbul, seni çok sevdim. Tekrar geleceğim!” She raises her hand and naturally covers the lens. AUDIO: Authentic Istanbul ambience — traffic, trams, ferries, seagulls, footsteps, vendors, cafés, wind and pedestrians — mixed with a subtle upbeat travel soundtrack. Natural Korean-accented Turkish dialogue. REALISM LOCK: Authentic Istanbul locations, realistic crowds, food, coffee, ferry movement, wind, shadows, reflections, skin, hair and fabric physics. Natural expressions and imperfect camera behavior. NEGATIVE: No face drift, duplicate people, distorted hands, floating objects, fake crowds, plastic skin, beauty filter, exaggerated reactions, impossible architecture, teleporting, subtitles, captions, text, logos, watermark, CGI or tourism-commercial aesthetic.

Cinematic scene

6 shots, total 30s 0.0-5.0s, 5.0-10.0s, 10.0-15.0s, 15.0-20.0s, 20.0-25.0s, 25.0-30.0s. Hard cuts, no dissolves. Normal speed throughout, no slow motion, no ramping. CAPTURE CADENCE CRITICAL: native 24fps, true 180-degree shutter, real 1/48s exposure every frame. Fluid, continuous motion blur. Never choppy, never juddering. No interpolation, no ghosting, no video look. NO ON-SCREEN TEXT CRITICAL: no text of any kind anywhere in frame. No captions, subtitles, titles, credits, watermarks, logos, timecode, UI overlays. NOBODY ELSE IN FRAME CRITICAL: no monks, villagers, or extras. Only the swordsman, the oni, and one crow are ever visible. THE FIRE CRITICAL: burning temple embers drift and flare through every frame, occasionally flaring bright enough to strobe-light a beat. The stepped quality of any lit moment comes entirely from ember-flare, never from broken footage camera motion stays smooth throughout. Falling ash rakes diagonally, screen-left to right. Subject Lock the swordsman: lean weathered build, closecropped greystreaked hair, hard jaw, sootstreaked skin, clean face, no tattoos. Charcoal hakama and a scorched indigo haori, one steel pauldron, straw sandals. Grips a curved katana two-handed. Advances through the burning temple courtyard toward the oni, stance low and controlled through every strike. Subject Lock the oni: towering horned demon, no visible pores cracked obsidian-red skin glowing along fault lines with molten orange light, tusked jaw, matted black mane. Carries a massive iron kanabo studded with spikes. Advances with heavy ground-shaking steps, never runs. World Plate: a burning temple courtyard at night a shattered stone torii gate, a collapsing pagoda roof shedding tile, a wide gravel yard scattered with broken lanterns, distant burning rooftops beyond the temple wall. No other structures stand whole. THE ATMOSPHERE CRITICAL, depth only: heavy smoke haze at high density swordsman sharp near, oni softened at mid-distance, burning rooftops near-erased far. Thickened smoke-air only, never a fog-machine look. SHOT 1 0.0-5.0s. THE ADVANCE. Low camera at gravel level, cant 15°, tracking the swordsman forward. Swordsman strides through drifting embers, katana raised, oni looming ahead. Swordsman screenleft advancing, oni screen-right filling the yard. Diegetic sound. SHOT 2 5.0-10.0s. THE CLASH. Camera orbits tight at chest height, cant swinging 15-35°, never square. Katana rings off the kanabo in a shower of sparks, swordsman thrown back a step. Both centered, gravel kicking up. Diegetic sound. SHOT 3 10.0-15.0s. THE OMEN. Cut to a crow bursting off the collapsing pagoda roof as tile showers down. It wheels hard away from the falling debris. Crow centered against the burning roofline. Diegetic sound. SHOT 4 15.0-20.0s. THE COLLAPSE. Low static camera in the courtyard, whip-panning up as the pagoda roof gives way. The roof section tears free and crashes into the yard; the swordsman rolls clear through gravel and ash. Roof beam diagonal across frame, swordsman emerging low center. Diegetic sound. SHOT 5 20.0-25.0s. THE SURGE. Camera drops to ground height, rockets forward through a corridor of drifting embers lit by a flare of firelight. Swordsman sprints low, katana trailing sparks, closing on the oni's flank. Swordsman screen-left driving right. Diegetic sound. SHOT 6 25.0-30.0s. THE AFTERMATH. Camera pulls back to a wide static courtyard frame, embers settling, wind dying. The swordsman stands alone, katana lowered, chest heaving, as the oni's glowing fault-lines dim to black and it topples out of frame. Swordsman small and centered, burning pagoda dark behind. Diegetic sound. Cross-Frame Rules: swordsman's haori, pauldron, and sandals never change. Oni's fault-line glow never fully fades until Shot 6. Only swordsman, oni, and crow ever visible. Ash direction stays constant. Pagoda keeps the same silhouette wherever it appears. Ember-flare applies only inside Shots 1, 2, and 5. No score enters at any point. Last Frame: the swordsman stands alone in the courtyard, katana lowered, embers drifting down around him, the oni's dark fallen shape at the frame's edge, the pagoda's broken silhouette burning low behind. No on-screen text, no logo, no watermark. Sound Bed: diegetic only — crackling fire, falling tile, katana ringing off iron, the oni's low guttural roar, gravel crunching underfoot, a crow's sharp cry. No score, no subtitles. Camera & Capture Realism: ~32mm/68° FOV, vintage 2x anamorphic with oval bokeh and streak flares off the firelight, shallow depth of field, color-negative rendition, fine grain. Violent handheld cant swinging 15-45°, punching in and ripping back, every frame mid-move but smooth in its own travel, never locked, never gimbalglide. No CGI look, no AI smoothness, no videogame HUD, no motion smoothing.

Cinematic scene

用 iPhone 拍摄的真实感短视频，9:16，15秒。不要电影感，不要炫光，不要体积光。自然平光，略微过曝，画面偏灰发白，像手机随手拍。真实纪录片质感，轻微手持抖动，偶尔失焦，运动模糊。【0-4秒｜拆弹】普通水泥地下室，顶部一盏廉价日光灯管，平淡冷白光，没有戏剧光影。一名穿旧防爆服的人俯身在一个锈金属箱上，箱上红色数字 00:06 跳动，红蓝线缠绕。手机怼脸拍，画质普通，背景虚化很浅，有噪点。剪线钳夹住红线。表情紧张但不夸张。【4-7秒｜反转】咔嚓——不是爆炸，一股白蒸汽"嘶"喷出。箱子弹开露出一台普通家用意式咖啡机。他愣住。手机慌乱晃动，自动对焦来回拉风箱。平光，无滤镜。【7-11秒｜咖啡】手机微距怼近白瓷杯，浓咖啡注入，crema油泡，热气。光线普通室内光，不是影棚。略微手抖，不完美对焦。真实。【11-13.5秒｜落点】他摘头盔，一脸疲惫平静地喝了一口，背后墙壁普通脏旧，没有god ray光束。日常平光。deadpan。【13.5-15秒｜品牌】切纯白背景，居中标志："PILA"红色粗体，下方 "SERIOUSLY GOOD COFFEE"，金色咖啡豆图形。关键：全片模仿手机随手拍，不要专业布光，不要高对比，不要湿润高光和CGI光泽。要平、要糙、要真实的皮肤毛孔和哑光布料。每个镜头只用一个运镜。

Cinematic scene

Cinematic fashion film of a young woman with long dark wavy hair walking through New York City. She wears oversized brown tortoiseshell sunglasses, gold hoop earrings, a cream sherpa fleece jacket with black leather trim and toggle buttons, a black mini skirt, black tights, and chunky black loafers, carrying a small black shoulder bag. Sequence: she exits a glass-door building, crosses a busy intersection with yellow taxis, glances into a café window, walks the sidewalk, buys a cold water bottle from a street cart, drinks it, sits on a green park bench, then walks away. Natural daylight, golden-hour highlights, smooth handheld and tracking shots, close-ups of her face, hair, and outfit details. Aesthetic street-style, 4K, filmic color grade, shallow depth of field.

一句話瞭解

這個模型表現如何？

純文字、首尾影格與 1～3 張參考圖片路徑會依已提交圖片選擇對應生成模式。直接生成 720p 扣 30 點數，1080p 扣 38 點數（37.5 無條件進位），4K 扣 90 點數。Extend 扣 30；符合條件的 1080p、4K 結果操作另扣 8、60 點數。

它擅長什麼: 目前定位是 Kyeo 的 Fast 公開資料接入頁，而不是 Google 官方 Veo 3.1 Fast 的完整鏡像。
適合哪些使用者: 適合用非關鍵素材逐一驗證文字、首尾影格或參考圖片路徑，並在基礎任務成功後另行評估結果操作。
為什麼要在 Kyeo AI 使用: Kyeo AI 將 Veo 的可用參數、點數消耗與結果操作放在一起，方便比較方案並立即開始創作。

關鍵參數總覽

協助你快速評估此模型是否符合你的業務情境。

模型類別

AI 影片模型

廠商

Google

模型家族

Veo

整合方式

目前站內整合

執行方式

非同步任務

提示詞上限

5,000 字元

上傳限制

JPEG / PNG · 1–3 · 10 MB

其他名稱

同一個模型在官方文件、社群討論與 API 串接層可能有不同稱呼，這裡幫你統一整理。

Veo3.1 Fast

Google Veo

Veo Video

使用者常見問題

從實際任務、輸入條件和結果要求出發，整理選擇模型時最需要確認的問題。

Veo3.1 Fast 怎麼用

Veo3.1 Fast REFERENCE_2_VIDEO 是什麼

Veo3.1 Fast 價格

Veo3.1 Fast 和 Quality 差異

Veo3.1 Fast 和 Lite 差異

Veo3.1 Fast 支援幾張參考圖

Veo3.1 Fast Extend 是什麼

Veo3.1 Fast Get 1080P 是什麼

Veo3.1 Fast Get 4K 是什麼

Kyeo AI 的 Veo3.1 Fast 與 Google 穩定版 Fast 有什麼區別？

選擇指南

還在猶豫該用哪個模型嗎？先查看這幾個關鍵決策點。

什麼情況可以先測試

當你需要驗證 Kyeo 目前 Fast 流程中的一條明確路徑，並願意先用非關鍵素材確認請求、回呼、結果與扣點時，可以從最小案例開始。

什麼時候先比較其他模型

若只需要較低的 Kyeo AI 扣點，可比較 Lite；Sora 2 Pro 已屬 Legacy，不適合作為預設交付基準。

最小驗證路徑

最小驗證路徑: 交付前請以非敏感樣例核對生成模式、時長、實際像素、音訊、網址期限與實際扣點。

使用者常搜尋的比較

把常見候選放在同一任務下比較，協助你確認輸入方式、控制項和成本差異。

Veo3.1 Fast 和 Veo3.1 Lite 差異

Veo3.1 Fast vs Veo3.1 Lite

Veo3.1 Fast 和 Veo3.1 Quality 差異

Veo3.1 Fast vs Veo3.1 Quality

Veo3.1 Fast 和 Sora 2 Pro 差異

Veo3.1 Fast vs Sora 2 Pro

模型比較表

還在猶豫該選哪一個嗎？這張表協助你快速看清目前模型與替代方案的核心差異。

Kyeo 目前角色

Veo3.1 Fast

包含三種輸入路徑與結果操作的站內 Veo 工作流程

Veo3.1 Lite

基礎生成輕量檔

Veo3.1 Quality

高品質檔

生成模式

Veo3.1 Fast

文字 + 首尾幀 + 參考圖

Veo3.1 Lite

基礎文生 / 圖生

Veo3.1 Quality

基礎文生 / 圖生，不開參考圖模式

結果操作入口

Veo3.1 Fast

成功任務後顯示 Extend / 1080p / 4K

Veo3.1 Lite

目前不支援這些後續動作

Veo3.1 Quality

同樣支援結果動作

本地基礎扣點

Veo3.1 Fast

固定 30 點數

Veo3.1 Lite

固定 15 點數

Veo3.1 Quality

固定 225 點數

目前驗證重點

Veo3.1 Fast

先確認一條基礎路徑與任務狀態閉環

Veo3.1 Lite

低門檻試樣

Veo3.1 Quality

目前驗證重點 — Veo3.1 Quality: 目前可用的同類模型，可從詳情頁繼續比較功能與價格。

實際使用體驗

結合公開資料、目前站內邊界和典型任務，說明模型適合與不適合的情況。

Veo：快速試用亮點

Veo：快速試用亮點: 本頁描述目前已驗證的站內合約，不公開也不依賴內部供應商模型識別碼。

基礎請求與結果操作必須分開計價

基礎請求與結果操作必須分開計價: 直接生成 720p 扣 30 點數，1080p 扣 38 點數（37.5 無條件進位），4K 扣 90 點數。Extend 扣 30；符合條件的 1080p、4K 結果操作另扣 8、60 點數。

功能詳解

目前站內的三種輸入路徑

目前站內的三種輸入路徑: 純文字、首尾影格與 1～3 張參考圖片路徑會依已提交圖片選擇對應生成模式。

參考圖片模式是 Kyeo 的產品選擇

參考圖片模式是 Kyeo 的產品選擇: 參考圖片模式接受 1～3 張經簽章驗證的 JPEG 或 PNG，只支援 16:9／9:16，且時長必須為 8 秒。

結果操作依賴站內原始任務

結果操作依賴站內原始任務: 升解析度操作只接受本人成功的 720p 基礎任務或 Extend 結果；直接生成的 1080p、4K 結果不能再次要求升解析度。

適合做什麼

Veo：日常創作

Veo：日常創作: 交付前請以非敏感樣例核對生成模式、時長、實際像素、音訊、網址期限與實際扣點。

參考圖驅動影片

參考圖驅動影片: 參考圖片模式接受 1～3 張經簽章驗證的 JPEG 或 PNG，只支援 16:9／9:16，且時長必須為 8 秒。

成功任務的結果操作驗證

成功任務的結果操作驗證: 升解析度操作只接受本人成功的 720p 基礎任務或 Extend 結果；直接生成的 1080p、4K 結果不能再次要求升解析度。

提示詞技巧

先選對 `generationType`，再最佳化 prompt

Veo Fast 目前最重要的不是多寫幾個風格詞，而是先判斷你要文字、首尾幀，還是參考圖模式。路徑選錯了，prompt 再長也很難補回來。

參考圖模式先把圖片職責分清楚

使用 1～3 張參考圖時，先標註每張圖負責主體、造型或風格中的哪一項，再逐項檢查是否混用或遺漏。

把升頻與延長視為獨立步驟

兩者都是成功任務後的獨立操作與額外扣點。解析度提高不代表內容、動作或細節一定更好。

選擇它的理由

純文字、首尾影格與 1～3 張參考圖片路徑會依已提交圖片選擇對應生成模式。 4 / 6 / 8 s.

參考圖片模式接受 1～3 張經簽章驗證的 JPEG 或 PNG，只支援 16:9／9:16，且時長必須為 8 秒。 1–3 JPEG / PNG.

直接生成 720p 扣 30 點數，1080p 扣 38 點數（37.5 無條件進位），4K 扣 90 點數。Extend 扣 30；符合條件的 1080p、4K 結果操作另扣 8、60 點數。 30 / 38 / 90.

升解析度操作只接受本人成功的 720p 基礎任務或 Extend 結果；直接生成的 1080p、4K 結果不能再次要求升解析度。 720p / Extend.

使用前須知

參考圖片模式接受 1～3 張經簽章驗證的 JPEG 或 PNG，只支援 16:9／9:16，且時長必須為 8 秒。 Auto.

升解析度操作只接受本人成功的 720p 基礎任務或 Extend 結果；直接生成的 1080p、4K 結果不能再次要求升解析度。 1080p / 4K.

音訊能力仍屬實驗性且不保證可用；交付前必須逐項人工檢查結果。升解析度操作只接受本人成功的 720p 基礎任務或 Extend 結果；直接生成的 1080p、4K 結果不能再次要求升解析度。

交付前請以非敏感樣例核對生成模式、時長、實際像素、音訊、網址期限與實際扣點。 10 MB.

可調整參數

協助你快速評估此模型是否符合你的業務情境。

生成模式

可選

參數類型: select

依已提交圖片決定

TEXT_2_VIDEO

FIRST_AND_LAST_FRAMES_2_VIDEO

REFERENCE_2_VIDEO

寬高比

可選

參數類型: select

預設值: 16:9

橫向 (16:9)

直向 (9:16)

自動裁切 (Auto)

解析度

可選

參數類型: select

720p

720p — 30

1080p — 38

4K — 90

時長

可選

參數類型: select

8 s

4 s

6 s

8 s

浮水印文字

可選

參數類型: text

提示詞翻譯

可選

參數類型: select

預設值: true

自動翻譯為英文

保留原文

點數消耗

依解析度為 30 / 38 / 90 點數

直接生成 720p 扣 30 點數，1080p 扣 38 點數（37.5 無條件進位），4K 扣 90 點數。Extend 扣 30；符合條件的 1080p、4K 結果操作另扣 8、60 點數。

省點數小技巧

批次生成圖片或影片前，建議先用同一組素材，在目前模型與同類候選模型之間進行 A/B 測試；確認效果後再批次生成，避免浪費點數。

常見問題

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。

Veo3.1 Lite

Kyeo AI 的 Veo3.1 Lite 支援純文字、首尾影格與參考圖片生成，可直接選擇 720p、1080p、4K，以及 4、6、8 秒時長。依解析度為 15 / 23 / 75 點數.

AI 影片模型

依解析度為 15 / 23 / 75 點數

查看詳情立即使用

Veo3.1 Quality

Veo3.1 Quality 是 Kyeo 目前採用 Veo 3.1 Quality 公開資料識別碼的 Veo 路線，純文字或 1～2 張首尾影格圖片的基礎請求固定 225 點數；成功後可另行發起 Extend、1080p 或 4K。Quality 是工作流程名稱，不代表每次結果都能達到固定的交付品質。

AI 影片模型

依解析度為 225 / 233 / 285 點數

查看詳情立即使用