SuperAPI AI Directory
Partner with us中文AI newsFavorites

AI video

2,117 results · Page 41 of 59

Grok Imagine

AI video

Grok Imagine is a multimodal AI image and video generation tool launched by xAI, powered by the Aurora engine. It supports text-to-image, text-to-video, and image-to-video generation, and can output synchronized audio in the same generation. Users can upload up to 9 images, 3 videos (total duration not exceeding 15 seconds), and 3 audio files, and use natural language to reference the motion, effects, shots, characters, and scenes within them. Generated videos range from 1 to 30 seconds in duration and support aspect ratios such as 16:9, 9:16, 4:3, 3:4, 21:9, and 1:1, with a maximum of 720p (other video models in the same workspace can reach 1080p). Registration grants free credits, and daily check-in allows manual claiming of 5–10 credits; paid plans are billed annually, and all paid plans include commercial licensing, watermark-free video downloads, and more than 20 AI models.

Read more

Grok Imagine

AI video

Grok Imagine is an AI video generation tool powered by the xAI Aurora engine. It supports text-to-video, image-to-video, and automatically generates synchronized background music and sound effects. It offers three creation modes: Normal, Fun, and Spicy, as well as five image and video aspect ratios: 1:1, 2:3, 3:2, 9:16, and 16:9. Generated videos range from 6 to 30 seconds in length, and the official claim is that it can be completed within seconds. The steps to use it are: enter a text prompt or upload an image, select a mode and aspect ratio, click generate, and download the video with audio. Paid plans are billed annually and are divided into three tiers: $143.3/year including 1,000 credits per month, $287.3/year including 2,400 credits per month, and $575.3/year including 6,000 credits per month; the official website also provides an entry point to start with free credits. Image-to-video only supports Normal and Fun modes.

Read more

Grok Imagine 2.0

AI video

Grok Imagine 2.0 is a standalone online AI video generator that can be used without an xAI subscription. It combines text-to-image, image editing, text-to-video, and image-to-video capabilities, and can turn text prompts into cinematic, anime, commercial, UGC, story, or VFX-style videos, and also supports uploading images to animate them. The platform provides motion control, multiple video styles, HD output, and a developer API, with generation times of about 30 to 90 seconds. According to the page, users can try it with free credits after registering, can start creating without a credit card, and can download videos in MP4 format for social media, marketing, or creative projects.

Read more

Hailuo 03

AI video

Hailuo 03 is a browser-based AI video generation tool powered by the Minimax Hailuo 3.0 model, offering text-to-video and image-to-video capabilities, and claiming to generate results in about 2 seconds, with typical processing times of 15-30 seconds. The tool supports high-definition resolution and 4K output, can maintain video character consistency, and is suitable for series content, brand visuals, advertising, and social media videos. Users can generate by entering prompts or uploading images, and can refine adjustments with natural language instructions. Free users initially receive 2 credits, and each generation consumes 2 credits; subscription plans provide credits monthly, with specific prices not yet provided.

Read more

Hailuo03AI

AI video

Hailuo03AI is an online AI video generation tool based on the MiniMax H3 model. It can generate 2K videos in the browser through text prompts or first-frame images, with selectable durations of 5 seconds, 10 seconds, or 15 seconds, and supports 16:9, 9:16, 1:1, and source image aspect ratios. The website provides official examples and prompt references, generation task tracking, failure refund explanations, and subscription plans such as Starter and Pro billed by credits.

Read more

Happy Horse

AI video

Happy Horse is an AI video generation tool that can combine text, images, video clips, and audio as input to generate cinematic videos with native sound effects, a multi-shot edited feel, and realistic motion in a single pass. It emphasizes multimodal input, realistic physics and motion performance, characters and scenes that remain consistent across frames, cinematic camera movement, and native audio synchronized generation, making it suitable for creative scenarios such as advertising and brand content, education and training, short film storytelling, social media short videos, dance and fitness, e-commerce product showcases, travel, fashion, and music visualization. The official website provides upload entry points for images and audio/video, supports generation via natural language prompts with references to specified materials, and after generation allows extending, editing clips, or iterative adjustments. The page also mentions a limited-time 50% discount for annual payment and an entry point for FAQ inquiries.

Read more

Happy Horse 1.0

AI video

Happy Horse 1.0 is an open-source AI video generator, billed as a SOTA open model, supporting video generation from text or images, with native synchronized audio generation. At its core is a 15-billion-parameter, 40-layer self-attention unified Transformer, employing DMD-2 distillation technology to require only 8 denoising steps, generating 1080p video in about 38 seconds on an H100. It supports ultra-low WER lip-sync in 7 languages: English, Mandarin, Cantonese, Japanese, Korean, German, and French, and integrates native joint audio-video synthesis of dialogue, ambient sound, and Foley. The model is fully open-source, including base models, distilled models, super-resolution modules, and inference code, allowing self-hosting and fine-tuning, and granting commercial usage rights. The site offers one-time purchase credit packages, credits never expire, and new users receive free credits to try it out. It is suitable for multilingual video production scenarios such as filmmaking, content creation, marketing videos, and e-commerce promotion.

Read more

Happy Horse 2

AI video

Happy Horse 2 is a web-based AI video generation studio that turns photos and text prompts into short films with a cinematic feel. Users can choose engines such as Veo 3.1, Sora 2, or Seedance 1.5 Pro, set the aspect ratio and duration, and then download the finished video. It supports image-to-video, reference images, first-and-last frames, and other modes, and you can also first generate a starting still frame in image mode using a variety of image engines before converting it to video.

Read more

Happy Horse AI

AI video

Happy Horse AI is a free AI video generation tool powered by the Happy Horse 1.0 model. The model is a single self-attention Transformer capable of jointly generating video and audio from text or image prompts, without the need for post-production dubbing or audio-visual synchronization. It supports two modes, text-to-video and image-to-video, and can generate multi-shot videos up to 15 seconds long in a single pass, automatically producing synchronized speech, spatial sound effects, and precise lip sync. The official description states that it offers fast inference, supports multiple languages, and is fully open source. Users can create videos through a three-step process: describe the vision, generate and refine, and preview and download high-definition output. The official website provides a free entry point and also lists three subscription plans—Basic, Standard, and Premium—which include different amounts of credits, video and image generation quotas, and commercial use licenses, among other things.

Read more

HappyHorse

AI video

HappyHorse 1.0 is an open-source AI video generation model that can simultaneously generate 1080p video and synchronized audio in a single inference pass. It adopts a unified audio-video architecture, supports text-to-video and image-to-video, supports native 1080p/2K output, and natively supports lip-sync in 7 languages: Mandarin, Cantonese, English, Japanese, Korean, German, and French. The official website states that it ranks first in blind-test voting on the Artificial Analysis Arena, and achieves 8-step fast inference through DMD-2 distillation. Users can upload reference images in JPG, PNG, and WEBP formats (≤50MB). Generation takes about 5–9 minutes; billing is by credits consumed (480p is 6 credits/second, 720p is 12 credits/second), with monthly, annual, and one-time credit packages available.

Read more

HappyHorse 1.0

AI video

HappyHorse 1.0 (Happy Horse AI) is an AI video generation tool that transforms text prompts or images into cinematic videos, supporting clips of up to 30 seconds at 1080p or 4K resolution. After registering, users receive 20 free credits with no credit card required. It supports bilingual English and Chinese prompts, offers multiple aspect ratios (portrait, landscape, square), and claims that users have full commercial rights to the generated videos. The platform uses credit-based pricing, offering monthly subscriptions, annual plans (save up to 50%), or one-time credit packages, and can be used without any video editing experience.

Read more

HappyHorse AI

AI video

HappyHorse AI (Happy Horse AI) is an AI video generation model released in 2026, built on a single-stream 40-layer Transformer architecture with approximately 15 billion parameters. It supports text-to-video and image-to-video generation, and can integrate audio-video generation, natively outputting 1080p HD quality, with an average generation time of about 10 seconds. It supports combined control of up to 12 types of multimodal inputs (text prompts, reference images, reference videos, audio references), covering lip sync, ambient sound, multilingual dubbing, and offers more than 50 visual styles. The product is divided into HappyHorse Lite (includes initial credits, allowing quick trial) and HappyHorse 1.0 (paid unlock of full 1080p, faster generation, and advanced multimodal features). It uses a credit-based billing system, offers annual payment and one-time credit purchase plans, and open-sources the base model, distilled model, super-resolution module, and inference code.

Read more

HappyHorse AI

AI video

HappyHorse AI is a browser-based AI video creation platform built on the Happy Horse (Happy Horse 1.0) model. Happy Horse was developed by the Future Life Lab team at Alibaba ATH (Taotian) and released in April 2026. It uses a unified 15-billion-parameter Transformer with DMD-2 distillation, generating video and synchronized audio simultaneously in a single forward pass, and supports phoneme-level lip sync in 7 languages. According to its introduction, Happy Horse ranked first in both text-to-video (Elo 1333) and image-to-video (Elo 1392) in blind tests on the Artificial Analysis Video Arena. HappyHorse AI lets users use the model directly in the browser to complete text-to-video, image-to-video, reference-to-video, audio, lip sync, and editing without downloading, without a GPU, and without waiting for open-source weights, and it supports 1080p output (up to 2K via super-resolution). The platform also offers free AI image generation without login. Pricing includes monthly, annual, and limited founder seats. The official GitHub and Hugging Face weight links currently show "coming soon".

Read more