Search AI Products and News

Memories.ai secures $8M funding for AI that analyzes 10M hours of video, surpassing 1-2 hour limits. Targets security/marketing with noise reduction, compression, and indexing. Faces competition but leads in scalability and edge computing.....

NTU's 'WORLDMEM' model enhances virtual reality with long-term scene consistency via innovative memory mechanisms and conditional diffusion transformers.....

Recently, ByteDance's Seed team officially launched a new Vision-Language-Action Model (VLA) called GR-3. This model demonstrates breakthrough capabilities in the field of robotic manipulation, not only understanding language instructions that include abstract concepts, but also precisely handling flexible objects. It also has the ability to generalize quickly to new tasks and recognize new objects. This achievement is seen as an important advancement toward a 'general-purpose robot brain'. Traditional robotic manipulation models often rely on large amounts of robotic trajectory data for training.

Recent research questions the mathematical ability of Alibaba's Qwen2.5 model, suggesting that its high scores may stem from data memorization rather than genuine reasoning. Tests show that Qwen2.5's performance drops sharply on previously unseen clean benchmarks; however, it can accurately reconstruct 60% of known questions, indicating training data contamination. The completion rate in the new LiveMathBench test dropped to zero. The study warns that contaminated benchmarks can mislead AI progress evaluation and emphasizes the need for clean benchmarks and multi-model validation. This finding raises doubts about the effectiveness of current AI evaluation methods and highlights the issue.

Lightricks releases open-source LTX-Video13B, a 13B-parameter video generation model with multi-scale rendering, achieving 30x faster speeds. It runs on consumer GPUs, supports 1216×704 real-time generation, and offers text/image/video-to-video modes. The model enhances coherence and detail, enabling keyframe control and style transfer. Free for SMEs, it includes training tools and optimized versions to democratize AI video creation.....

NVIDIA's Canary-Qwen-2.5B sets a 5.63% WER record on Hugging Face OpenASR. This CC-BY licensed model combines FastConformer encoder with Qwen3-1.7B LLM decoder for efficient speech-to-text and NLP. Supports multi-GPU deployment for cloud/edge applications.....

Google launched Veo3 API for text-to-video & audio generation, but costs are high ($0.75/s for 720p). Used in gaming cinematics & 3D modeling. Faster 'Veo3Fast' mode not yet API-accessible. Requires Gemini API & Google Cloud subscription.....

Meta and Berkeley developed StreamDiT for real-time AI video generation: 1) 16fps 512p on single GPU, 4B-param model creates 1-min videos with live edits; 2) Novel buffer enables parallel processing (2 frames/0.5s) with 8-step optimization; 3) Trained on 3K HD videos + 2.6M dataset; 4) Outperforms rivals in motion smoothness, 30B-param version shows quality potential; 5) Enables interactive video despite transition flaws.....

Modo AI introduces a 30-second rapid prototype generation feature, supporting multi-device adaptation and conversation optimization. Users can generate high-fidelity, editable prototypes through text, sketches, and other input methods, and support iterative conversation adjustments. The AI can intelligently parse uploaded sketches, wireframes, and more, automatically generating interfaces. It offers dual-mode editing, automatic documentation generation, and code integration features, covering multiple scenarios such as e-commerce and social networking, significantly lowering the barrier to prototype creation and improving product design efficiency.

Tencent launched Hunyuan3D-PolyGen, the industry's first art-grade 3D generation model, using self-developed BPT technology to enhance wiring quality and complex object modeling. It generates high-precision geometric models, supports multiple surface types, and boosts gaming pipeline efficiency by 70+%.....

As competition in the field of artificial intelligence intensifies, the CEO of OpenAI, Sam Altman, recently had a phone conversation with the CEO of Microsoft, Satya Nadella, discussing their future cooperation. This information was revealed in an interview on Altman's podcast on Tuesday. He mentioned that the conversation with Nadella mainly focused on how to revise their investment terms and issues related to future equity. It is known that Microsoft is a key investor in OpenAI, and recently there have been some differences between the two sides regarding investment details, especially concerning Microsoft's future shareholding.

Shoplauncher AI is revolutionizing the barrier to entry for independent store creation through AI technology, promising a fully optimized Shopify store in minutes. This review will analyze how it addresses the three major obstacles of technology, product selection, and optimization, and challenges faced by 85% of novice sellers may be easily resolved.