AI News | Latest AI Breakthroughs, AI Trends & AI Policy Updates – AGIYes
OpenAI has officially launched its newest voice model, GPT-Realtime, an advanced multimodal voice agent that is quickly gaining industry attention. Designed to deliver powerful reasoning, image input capabilities, and improved instruction adherence, this model sets a new benchmark for intelligent voice […]
OpenAI Launches GPT-Realtime: Supporting Image Input and Real-Time Calls Read More »
August 28, 2025, xAI officially released Grok Code Fast 1, a large-scale programming model designed for developers, codenamed “Sonic.” This model aims to solve the common pain points of slow inference and tool invocation in current programming assistants, providing developers
xAI’s Grok Code Fast 1 Release Makes Developers’ Programming Faster and Cheaper Read More »
Recently, Meta officially announced a major update for its virtual-reality social platform——Meta Horizon Worlds Update. The centerpiece of this update is a powerful set of AI tools that will let developers easily create and deploy “soulful” AI non-player characters. These
Meta Horizon Worlds Update Lets Developers Create AI-Driven NPCs Read More »
In a significant move within the AI-generated music space, Beatoven.ai has officially launched Maestro—a new foundation model for generative music creation. Unlike many AI music tools, Maestro emphasizes ethical sourcing and legal compliance through formal partnerships with rights holders. This
Recently, the AI-powered The Wizard of Oz, a collaboration between Sphere, Warner Bros., and Google, premiered, sending shockwaves through the global film industry. This project utilized artificial intelligence to completely digitally remake the 1939 classic, from its visuals to its
AI-powered The Wizard of Oz Premieres: A Head-on Collision Between AI Film and a Classic Read More »
Aishi Technology has officially launched its latest AI video generation model, PixVerse V5, marking a significant leap in AI-driven content creation. The release comes alongside a major company milestone: surpassing 100 million global users. Recent evaluations from the independent AI
August 27, 2025 — ByteDance’s digital human team released its multimodal digital human solution, OmniHuman-1.5. As an upgraded version of OmniHuman-1, the framework converts a single image and an audio clip into highly realistic dynamic video. The generated characters not
Google Translate, one of the world’s most-used translation tools, has recently received an unprecedented and significant update. This isn’t just a simple feature add-on; it’s a technological revolution driven by the Google Gemini large language model. The Google Translate Major
In August 2025, Los Angeles–based start-up Robomart officially unveiled its fifth-generation autonomous delivery robot, the Robomart RM5. The launch is not just another technical milestone; it is seen as a direct challenge to the current food-delivery model. According to the
On August 26th, Google DeepMind announced the all-new Gemini 2.5 Flash Image model (internal nickname “nano-banana”). This model not only inherits the “speed” and “efficiency” advantages of its predecessor, Gemini 2.0 Flash, but also achieves a qualitative leap in image
