SuperAPI AI Directory
Partner with us中文AI newsFavorites

AI video

2,208 results · Page 30 of 62

musubi-tuner

AI video

musubi-tuner is an open-source script repository maintained by developer kohya-ss, used to train LoRA (Low-Rank Adaptation) models on a variety of video and image generation architectures. The repository provides tools such as dataset configuration, cache precomputation, training, and inference in the form of scripts. The supported architectures include HunyuanVideo, Wan2.1/2.2, FramePack, FLUX.1 Kontext, FLUX.2 dev/klein, Qwen-Image, Z-Image, and MiniMax-H3, among others. According to official materials, this repository is an unofficial project and has no affiliation with the official repositories of the aforementioned architectures, and the project is under development and may introduce breaking changes. Installation methods are divided into two categories: pip-based and uv-based, with different instructions provided for Linux/MacOS, Windows, and Windows on ARM64. The project has enabled GitHub Discussions for community Q&A and technical exchange, and Issues for bug reports and feature requests.

Read more

My Winter Car Wiki

AI video

My Winter Car Wiki is a wiki resource site built around the game "My Winter Car," primarily offering a complete build guide for the Corris Rivett model, a database containing over 200 parts, survival tips, and video tutorials. The site aims to help players explore this winter car-themed game. At present, the only confirmable information comes from its official one-sentence introduction, with no further details provided regarding pricing, platform, or specific usage methods.

Read more

Nano Banana Video

AI video

Nano Banana Video is an independent AI image and video generation service that creates AI images and videos via text prompts or reference photos. It integrates models such as Nano Banana, GPT Image 2, Seedance, and Kling, offering capabilities including image generation, reference image editing, and text/image-to-video, and provides e-commerce sellers with workflows for Amazon product videos, template videos, batch video generation, and multi-platform aspect ratio export. The official description states you can start for free, get up and running within 30 seconds, video generation can be as fast as about 2.1 seconds, and it claims to offer watermark-free export and commercial licensing. The service declares itself to be an independent AI wrapper service with no affiliation, endorsement, or sponsorship relationship with any AI provider.

Read more

Official homepage

AI video

iptv-org pro is an independent public TV channel directory website whose data comes from the open-source iptv-org dataset. The directory includes 9,908 public broadcast television channel entries, covering 177 countries, and lists 12,099 broadcast sources. Users can search or filter by country and category to find channels, view metadata, official broadcast websites, and streaming technical attributes on channel detail pages, and open the listed public stream URLs in external players. The platform only provides an information index; it does not host, store, or transmit video or audio broadcasts, and it does not sell IPTV subscriptions or commercial playlists. All stream links point to external third-party public URLs, and stream accessibility depends on the original broadcaster.

Read more

Official homepage

AI video

Hyperdream is an AI video generator and an end-to-end AI filmmaking studio. It integrates screenwriting, style setting, AI character casting, storyboarding, video generation, and editing into a single platform, and maintains contextual continuity throughout the entire process from script to finished film, thereby keeping characters and visual style consistent across scenes. Users can start from a single prompt or a one-sentence logline to generate a structured script, lock in a unified cinematic look (lighting, color grading, film grain), create AI actors with consistent appearance and voice, generate shot-by-shot storyboards, and use virtual camera controls, shot blocking, and shot selection to generate video in a "directed" manner, finally assembling shots in a built-in non-destructive editor, adding sound, and exporting the finished film. It is officially described as aimed at independent filmmakers, content creators, marketers, and production teams, and suitable for scenarios such as short films, music videos, advertisements, and trailers.

Read more