可用视频 API
Kling 4.0
Kling 4.0 API on PoYo. Generate 3–30 second videos from text, first and last frames, or image/video references and subjects. Edit source videos at 720p or 1080p.
FLUX 3 Max
Create 5–20 second videos with FLUX 3 Max from text, one or two images, or up to ten positioned keyframes. Choose 720p or 1080p and generate native synchronized audio.
Vidu Q4
Vidu Q4 API on PoYo. Turn a start image or up to 12 reference images and 3 voice clips into 3-16 second cinematic videos with native audio, multi-shot storytelling and 540p to 4K output. From 9 credits ($0.045) per second.
OmniHuman 2.0
OmniHuman 2.0 API on PoYo. Animate one portrait with 4-300 seconds of speech or singing into a 720p talking video with precise lip-sync, natural emotion, gestures and multi-person scenes. 21 credits ($0.105) per output second.
Kling 4.0 Flash
Kling 4.0 Flash brings fast 720p video creation up to 20 seconds, native audio, and Omni Reference for text, image, and multimodal creative workflows.
MiniMax H3 Max
MiniMax H3 Max generates 5–15 second videos from text, first/last frames, or image, video and audio references at 480p, 768p and 1080p.
MiniMax H3 Max Turbo
MiniMax H3 Max Turbo generates 5–15 second videos from text or first/last frames at 480p, 768p and 1080p.
Gemini Omni 1.1 Flash
Gemini Omni 1.1 Flash API for text-to-video, first and last frame interpolation, reference image generation, and reference-video workflows.
Wan 3.0
Generate videos up to 30 seconds with Wan 3.0 using text, first and last frames, or multimodal references including images, video, audio, documents, and links.
Wan 3.0 Video Prime
Wan 3.0 Video Prime is Alibaba's Wan 3.0 tier optimized for faster turnaround, with multimodal image, video, audio, document, and public webpage references, videos up to 30 seconds, and optional generated audio.
Wan Animate 2
Alibaba's end-to-end character animation model for direct motion transfer, identity preservation, and text-driven viewpoint control. Coming soon to PoYo.
MiniMax H3 / Hailuo 03
MiniMax H3 (Hailuo 03) generates native 2K video from text, first/last frames, or image, video, and audio references.
FLUX 3
Black Forest Labs' FLUX 3 video family for text, images, first and last frames, source-video extension, positioned keyframes, and native audio generation.
Seedance 2.5
ByteDance Seedance 2.5 video generation with 4-30 second output, first and last frame control, up to 50 image, video, and audio references, optional synchronized audio, and 480p, 720p, or 1080p delivery.
Seedance 2.0 Mini
ByteDance Seedance 2.0 Mini video generation for lower-cost text-to-video, image-to-video, and multimodal reference workflows with 480p and 720p output.
Kling 3.0 Turbo
Kling 3.0 Turbo is the speed-optimized Kling 3.0 variant for high-volume text-to-video and image-to-video generation with multi-shot storyboarding. Standard (720p) and Pro (1080p) models available.
Omni Flash
Omni Flash API access for text-to-video, image-to-video, three-image reference fusion, and video-input generation workflows.
Grok Imagine Video 1.5
Grok Imagine Video 1.5 API for text-to-video, single-frame image-to-video, and 1–7 image reference-to-video generation with up to 1080p output.
Kling Avatar 2.0
Kling Avatar 2.0 creates audio-driven talking avatar videos from one reference image and one driving audio file, with Standard and Pro quality modes.
Happy Horse 1.1
Alibaba Happy Horse 1.1 video generation with text-to-video, image-to-video, and reference-to-video workflows.
Happy Horse
Alibaba Happy Horse 1.0 video generation and editing with text-to-video, image-to-video, reference-to-video, and video-edit workflows.
Seedance 2
ByteDance Seedance 2 video generation with text-to-video, image-to-video, multimodal reference-to-video, video-to-video, and audio-to-video workflows using image, video, and audio references with optional native audio.
Wan 2.7 Video
Wan 2.7 Video models on PoYo for text-to-video, image-to-video, reference-to-video, and edit-video workflows.
Sora 2 Official
OpenAI Sora 2 on PoYo with synced audio, improved physics, optional reference image input, and fixed 4-second, 8-second, 12-second, 16-second, and 20-second tiers.
Kling 3.0 Motion Control
Kling 3.0 Motion Control is a reference-driven motion transfer model that combines one character image and one source video with transparent per-second pricing.
Hailuo 2.3
MiniMax's Hailuo 2.3 video model for realistic human motion, expressive characters, and text-to-video or first-frame guided generation at 768p and 1080p.
Kling 1.6
Kling 1.6 Standard and Pro video generation on PoYo with text, first/last frame, and multi-image element reference workflows.
Kling 2.1
Kling 2.1 on PoYo provides Standard and Pro image-to-video modes with 5-second and 10-second clips, start-frame control, and optional end-frame control in Pro.
Kling 2.5 Turbo Pro
Kling 2.5 Turbo Pro is a flexible short-form video model with text-to-video, optional frame guidance, smooth motion, cinematic depth, and fixed 5-second and 10-second tiers.
Wan 2.2 Fast
Wan 2.2 Fast provides fast text-to-video and image-to-video generation with low-cost 480p and 720p tiers for quick iteration.
Wan 2.5
Wan 2.5 combines text-to-video and image-to-video generation with 5-second and 10-second output, synchronized audio support, and multiple size and resolution tiers.
Runway Gen-4.5
Runway Gen-4.5 is a high-fidelity video model focused on prompt adherence, cinematic motion, visual fidelity, and optional reference image guidance.
Kling 3.0
Kuaishou's advanced Kling 3.0 video model family with Standard, Pro, and native 4K generation, multi-shot storyboarding, synchronized audio, and element-based consistency.
Kling O3
Kling O3 video generation on PoYo with Standard, Pro, and 4K variants for text-to-video, image-to-video, and reference-to-video workflows.
Kling 2.6 Motion Control
Kuaishou's motion control model that transfers motion from reference videos to character images while maintaining identity and adapting environments.
Seedance 1.5 Pro
ByteDance's latest video model with synchronized audio generation, flexible aspect ratios, and enhanced motion control.
Wan 2.6
Alibaba's Wan 2.6 video generation family for text-to-video, image-to-video, and video-to-video with multi-shot 1080p output.
Grok Imagine
xAI's Aurora-powered visual AI for image generation and video creation with Fun, Normal, and Spicy creative modes.
Hailuo 02
MiniMax's #2 globally-ranked video model with NCR architecture, ultra-realistic physics, and 1080p cinematic output.
Seedance 1.0 Pro
ByteDance's #1 ranked video model with multi-shot storytelling, cinema-grade motion, and bilingual text-to-video generation.
Kling 2.6
Kuaishou's revolutionary video model that simultaneously generates visuals with synchronized dialogue, sound effects, and ambient audio in one pass.
Wan Animate
Alibaba's 14B-parameter character animation model that transfers motion from reference videos to static characters with exceptional identity preservation.
Veo 3.1
Google Veo 3.1，1080p/60 秒视频，支持多参考图、首尾帧插值、场景延伸至 148 秒。
Veo 3.1 Official
Official VEO 3.1 video generation with text, image, first/last-frame and reference workflows across fast, lite and quality tiers.
平台对比
poyo.ai vs FAL vs Replicate vs KIE
对比主流 AI 视频生成平台，为专业团队选择最佳方案。
| 功能特性 | poyo.ai | FAL | Replicate | KIE |
|---|---|---|---|---|
| 价格优惠 | 最高 8 折 | 标准价格 | 标准价格 | 标准价格 |
| 计费透明度 | 积分制，被拦截不收费 | 按次计费 | 按次计费 | 按次计费 |
| 服务稳定性 | 生产级稳定 QoS | 因模型而异 | 因服务商而异 | 因服务商而异 |
| API 访问 | 完整 REST + 文档 | REST | REST | REST |
| 在线体验 | 支持 | 支持 | 支持 | 支持 |
常见问题
常见问题解答
关于 AI 视频 API 的常见问题。
视模型而定。Veo 3.1 支持最长 60 秒视频，通过场景延伸可达 148 秒；Sora 系列专注于高真实感的短视频生成，Pro 版支持 15 秒高画质输出。
可以。Sora 2/2 Pro 支持原生音画同步生成，包括对白、音效和环境音。您也可以导出静音视频后自行添加音轨。
支持。上传参考图片即可保留构图生成动态视频。Veo 3.1 还支持首尾帧插值和多参考图风格控制。
每次调用消耗 API 卡片上标注的积分。积分永不过期，文生视频和图生视频接口统一计费，被拦截的提示词不收费。
Sora 输出默认包含 C2PA 溯源认证。平台也提供水印和安全过滤功能，满足合规需求。
可以。您可以在提示词中描述镜头语言（推拉摇移、环绕等）和场景结构。多镜头叙事可使用 Cameo 功能保持角色一致性。