Everything tagged video-generation, newest first. Tags come from the classifier reading each item's summary; 394 tags used 25 times or more have their own page.
This update highlights a rapid wave of AI releases across major labs, with standout progress in open-source world models, interactive video generation, and Google’s new multivariate time-series foundation model, all emphasizing increasingly practical, reusable systems and broadly released training resources.
This update highlights a wave of new open-source AI tools and robotics breakthroughs, including real-time interactive video world generation, 4D human reconstruction from video, compact speech synthesis, improved image and video editors, vision-capable models, and record-setting robots that push speed, jumping, and dexterity.
LTX 2.5 is the fastest open-source model with improved features like diffusion fidelity rendering and multi-shot video generation, offering high-resolution outputs and compatibility with existing Loras, while being installable for free offline use.
Miniax H3 is a powerful open-source video generator that runs locally with minimal VRAM, excelling in character consistency, world knowledge, and multimodal input, offering exceptional animation capabilities and free offline usage.
Seed Dance 2.5 and Minimax H3 are advanced multimodal video generators capable of transforming videos with text, images, and audio inputs, excelling in different scenarios like high-action scenes and detailed animations.
Recent advancements in open-source AI have enabled real-time 3D human animation, mobile video generation, precise facial control, and music decomposition, while OpenAI introduces a model designed to challenge other AIs.
Avataar AI offers a distilled video model service at a cost of $0.005 per second of video generation.
The AI landscape is rapidly evolving with major developments like Google's Gemini 3.5 Flash, Alibaba's Quen 3.7, and Perser's Composer 2.5, each offering significant advancements in speed, efficiency, and multimodal capabilities, while Google IO's upcoming releases promise groundbreaking AI agent platforms and video generation models.
Runway, an AI video generation startup, believes that video generation is crucial for developing world models and sees its outsider status as a strategic advantage.
Google's new AI video model, Gemini Omni, was accidentally leaked, showcasing impressive video generation capabilities that rival existing models like Sea Dance 2, sparking widespread interest and speculation about its potential impact.
Recent AI advancements include Google's faster open-source model, a state-of-the-art video generator, AMD-trained models, and innovative tools like RecGen for 3D object reconstruction from images, alongside new top-tier open-source image and voice models.
Recent AI advancements include Alibaba's video generator, an open-source image model for text posters, AI tools for software like Photoshop, recursive multi-agent systems, and a new video editing AI capable of detecting transitions, though some models like Happy Horse have underperformed expectations.
This week in AI saw significant advancements with new model releases from Anthropic and Alibaba, innovative video generation techniques, and ultra-efficient open-source models like Turnary Bonsai, alongside state-of-the-art developments in text-to-speech and humanoid robotics.
Recent AI developments include Anthropic's powerful Claude Mythos model with significant cybersecurity implications, new models from Alibaba, Meta, Deepseek, and ZAI, and advancements in video game generation and anime modeling.
Google Flow is an AI creative studio combining Google's VO, Nano Banana 2, and Gemini models for generating and editing images and videos, with various pricing tiers offering different credit allocations and costs for usage.
PI API offers access to Cance 2.0 for video generation with customizable quality and duration options, though pricing varies and includes a watermark removal fee.
LTX 2.3 is a significant upgrade to the open-source video generator, offering improved motion consistency, prompt understanding, and audio quality, while being capable of running on low VRAM and generating up to 20 seconds of 4K video.
This week saw the release of groundbreaking open-source AI tools, including video and image editors, real-time video generators, and compact AI models for edge devices, showcasing significant advancements in AI-driven media editing and enhancement.
Seed Dance 2.0 by Bite Dance is an advanced AI video generator that excels in creating high-action, coherent scenes with accurate physics and cinematic quality using reference images and frames.
Cling 3.0 introduces advanced video generation features, including a multi-shot capability that enhances cinematic quality by allowing customizable scenes and consistent character portrayal across shots.
Recent AI advancements include Nvidia's Earth 2 for fast, accurate weather forecasting, a new open-source video generator MOA with sound, and real-time interactive AI models, showcasing significant progress in AI capabilities.
Higfield is a user-friendly AI creative suite that simplifies media creation with features like Nano Banana Pro for quick, high-quality image and video generation, offering customizable aspect ratios and unlimited usage, making it a cost-effective alternative to traditional tools like Google Gemini.
Google AI Studio has introduced major upgrades, including VO3.1 for enhanced video quality and vertical video generation, transforming it into a powerful tool for creating AI-driven apps and automating workflows with state-of-the-art models like Gemini 3 Pro.
Recent advancements in AI technology include open-source video generators with multimodal inputs, a video face swapper, a tutoring agent with personalized learning, and a powerful model for video editing and creation, all offering enhanced capabilities and local operation.
LTX2 is a cutting-edge, open-source video generator with built-in audio, 4K resolution, low VRAM requirements, and comprehensive features like ControlNet and Loras for enhanced video creation.
Stable Video Infinity version 2 Pro is a free, open-source tool that generates seamless, coherent long videos using low VRAM, offering significant improvements over previous versions in video quality and consistency.
Recent advancements in open-source AI models have introduced state-of-the-art capabilities rivaling top closed models, including video generation, image editing, photo focus adjustment, 3D scene creation, and autonomous video game playing, with notable releases like Nvidia's Nitrogen and Alibaba's Flashportrait.
This week has seen groundbreaking advancements in AI technology, including new video and image generators, a top-tier real-time text-to-speech tool, and impressive humanoid robot demos, all of which are accessible on consumer-grade hardware.
Cling has released two powerful AI video models, Cling01 and Cling 2.6, offering advanced features like background replacement, character insertion, and native sound integration, revolutionizing video editing with simple prompts.
Hunyan Video 1.5x10cent is an advanced, open-source AI video generator excelling in realistic motion, cinematic aesthetics, and versatile camera controls, especially suitable for low VRAM systems.
This week saw significant advancements in AI with new open-source video generators, Google's Gemini 3, Nano Banana Pro, XAI's Grock 4.1, OpenAI's coding model, and Depth Anything 3, which excels in 3D mapping from minimal input.
LTX2 is an open-source AI video generator capable of creating 4K videos with audio, offering features like text-to-video and image-to-video, and supports up to 20-second video generation, providing high-quality, realistic outputs with a comedic edge.
Recent developments in AI include advanced video generation models, autonomous gaming agents, and versatile image editors, showcasing significant advancements in video, gaming, and design technologies.
The new AI video generator, Hyo 2.3, excels in high-action scenes, physics, and complex prompt understanding, outperforming competitors Sora 2 and Vo 3.1 despite minor imperfections.
Google's VO3.1 AI video generator offers enhanced realism, richer audio, and improved prompt adherence, but is limited to 8-second clips, with features like ingredients to video and multiple shot generation.
OVI is a free, open-source AI video generator that creates videos with audio, including dialogue and sound effects, using text prompts, and supports multiple languages and character actions.
This week saw the release of several groundbreaking AI models, including video generators with transparency, real-time 3D head creation, and a new open-source text-to-speech model, alongside updates from major AI developers like Enthropic and Alibaba.
Sora 2 is a powerful video generator with advanced physical realism, audio integration, and diverse style capabilities, excelling in complex scenes, anime, and AI-generated cameos.
OpenAI's Sora 2 is a groundbreaking video generation model with enhanced realism, native audio, and innovative features like Cameo, revolutionizing creative possibilities.
Recent advancements in AI include new video generators, upgraded AI models from Google and Alibaba, and impressive humanoid robot demos, showcasing open-source tools and innovative features for creating and editing multimedia content.
The evolution of video generation models, like Google's V3, now creates high-fidelity images from noise using diffusion techniques, transforming vague concepts into detailed visuals through iterative noise reduction.
Google Photos is enhancing its photo-to-video feature by integrating the advanced video-generation model, Veo 3, for improved video creation.
Recent AI advancements include new video generators, 3D model editing tools, deepfake technology, and improved text-to-speech systems, showcasing significant progress in AI capabilities.
OpenAI has released two open-source AI models, GPT OOS 120B and GPOS 20B, offering advanced capabilities and performance, while other AI advancements include a 3D world generator, improved video and image generators, and Enthropic's new model, Claude Opus 4.1.
OpenArt has introduced a new feature designed to assist creators in producing AI-generated brain rot videos, enhancing creative possibilities.
Juan 2.2 by Alibaba is a powerful, open-source AI video generator excelling in cinematic video creation, offering offline use with low VRAM, and supporting complex prompts for detailed video outputs.
AI systems have advanced in generating videos from text prompts using diffusion models, which are deeply connected to physics and Brownian motion, enabling realistic video creation through iterative noise transformation and the integration of models like CLIP for enhanced image-video generation.
This week in AI, a new open-source video generator called Pussa, five times faster than its predecessor, introduces real-time, flexible video creation, alongside advancements in 3D modeling, autonomous robotics, and the debut of ChatGpt Agent.
Google has started deploying its Veo 3 video generation model to Gemini users across over 159 countries.
Recent AI advancements include an open-source text-to-speech generator with emotion control, Google's Gemini 2.5 Pro for game scene generation, Shape LLM Omni for 3D model creation, and Flow Mo for enhancing video generation quality.
