AI News
Daily curated AI news: breakthroughs, product launches, industry trends and more.
Latest News
Total 157 articles
AI News (2026/3/31): AiShi Tech Launches AI Video Model PixVerse V6
AiShi Tech launched the latest version of its AI video generation model, PixVerse V6, on the first day of the “AI Lightning Release Week.” The new version has been comprehensively upgraded in terms of character realism, complex motion, physical simulation, and audio-visual synchronization, with a maximum generation duration of 15 seconds. PixVerse V6 focuses on optimizing skin texture and emotional expression, enhancing stability in high-speed motion scenes, and supporting motion inertia and light continuity between shots, significantly lowering the creation threshold.
AI News (2026/3/28): Google Launches Gemini 3.1 Flash Live Real-Time Voice Model
Google has recently released the Gemini 3.1 Flash Live real-time voice model, its highest-quality tool for real-time voice processing to date. The model is now available in the Gemini App, Search Live, and Google AI Studio, featuring several core upgrades, including vibe coding, a context window that has doubled in size, and support for real-time interactions in over 200 countries. In the ComplexFuncBench audio test, the function call accuracy reached 90.8%, significantly outperforming its predecessor.
AI News (2026/3/27): Google Launches Gemini 3.1 Flash Live Real-Time Voice Model
Google has recently released the Gemini 3.1 Flash Live real-time voice model, which is currently its highest-quality real-time voice processing tool. The model is now available in the Gemini App, Search Live, and Google AI Studio, with core upgrades including vibe coding, a context window that has doubled in size, and support for real-time interactions in over 200 countries. In the ComplexFuncBench audio test, the function call accuracy reached 90.8%, significantly outperforming its predecessor.
AI News (2026/3/26): Google Launches AI Music Generation Model Lyria 3 Pro
Google has recently launched its latest AI music generation model, Lyria 3 Pro, which has seen significant improvements in generating music structure and length. Lyria 3 Pro can accurately handle elements such as intros, verses, choruses, and bridges, and supports the generation of complete audio tracks up to about 3 minutes long. Additionally, the model does not directly mimic specific artist styles, and its training data only uses legally authorized content, with all generated audio embedded with SynthID digital watermarks.
AI News (2026/3/25): OpenAI Announces Shutdown of Sora, Shifts Focus to Enterprise...
OpenAI has announced the shutdown of its video generation platform Sora, including the Sora App, API, and video functionality in ChatGPT. This move is part of a strategic realignment to prepare for an IPO, redirecting computational resources to the next-generation model “Spud” and enterprise productivity tools. Additionally, the three-year IP licensing agreement and $1 billion investment intent with Disney have been terminated.
AI News (2026/3/24): Alibaba DAMO Academy Unveils the Xuantie C950: The World's Most...
Alibaba DAMO Academy has launched the next-generation flagship RISC-V CPU IP, the Xuantie C950, which scores over 70 points in the SPECint2006 benchmark, making it the world's most powerful RISC-V CPU. The product is the first to achieve native smooth operation of large models with hundreds of billions of parameters (Qwen3, DeepSeek V3), integrating a 4K ultra-wide Vector engine and Matrix engine, with a single-core computing power of 8TFLOPS.
AI News (2026/3/23): MiniMax Token Plan: The World's First Subscription Plan Supporting...
MiniMax has launched the world's first subscription plan, Token Plan, that supports all modalities, including video, voice, music, and images. This plan allows a single API Key to meet the needs of code writing, content creation, and video generation. Additionally, the voice/video resource packages can save users 20% on costs.
AI News (2026/3/20): Cursor Launches AI Programming Model Composer 2
Cursor has officially launched its latest AI programming model, Composer 2, which is now available on the Cursor platform. Composer 2 has achieved significant improvements in coding capabilities, performance, and cost, making it a new choice for developers and tech enthusiasts.
AI News (2026/3/19): Xiaomi Launches Three Large Models for the Agent Era
Xiaomi recently released three large models for the Agent era: MiMo-V2-Pro flagship base model, MiMo-V2-Omni multimodal Agent base model, and MiMo-V2-TTS text-to-speech model. These models are being anonymously tested on the OpenRouter platform under the codenames "Healer Alpha" and "Hunter Alpha," aiming to advance AI technology in multimodal interaction and complex task handling.
AI News (2026/3/18): OpenAI Launches GPT-5.4 mini and GPT-5.4 nano, Performance Approaches...
OpenAI has recently released two lightweight models, GPT-5.4 mini and GPT-5.4 nano. GPT-5.4 mini achieved a score of 54.4% on the SWE-Bench Pro coding benchmark, 3.3 percentage points lower than the full version of GPT-5.4, but it runs twice as fast as the previous generation and supports up to 400,000 tokens of context. GPT-5.4 nano is positioned for ultra-lightweight tasks and costs only 1/12th of the full version.
Subscribe to Daily AI News
Get curated AI news delivered to your inbox daily. Never miss an update.
We never share your email. Unsubscribe anytime.