1841 articles
Indian startup Sarvam AI launched its first Edge AI smart glasses capable of working offline, with India's Prime Minister publicly testing the device at Bharat Mandapam.
Grokipedia, the Grok-reviewed knowledge platform, has crossed 701,974 approved edits across more than 6 million articles — and it's only getting started.
Researchers have developed a diffusion-based technique that trains on internal language model activations to stabilize AI outputs while allowing precise behavior steering, marking a breakthrough in controllable AI systems.
MiniMax launches a high-performance large language model with dramatically lower operating costs, intensifying pricing competition in the AI market.
A decentralized AI data network proposes token-backed expert validation across multiple domains through specialized prediction markets.
A newly released open-source image-editing AI model demonstrates leading results on editing benchmarks and runs locally under an Apache-2.0 license.
Alibaba released Qwen3.5-397B-A17B, the first open-weight model in the Qwen3.5 series, featuring multimodal capabilities and up to 19× faster decoding throughput than previous versions.
Kilo Code reclaimed the top spot on OpenRouter's daily leaderboard after processing 313 billion tokens, with the surge primarily fueled by heavy usage of Z.ai's newly released GLM-5 model.
Multiple flagship AI models are rolling out across leading platforms in the coming weeks. The packed release schedule signals how development timelines are tightening across the industry.
0G Labs is building decentralized AI computing infrastructure where every operation can be cryptographically verified, eliminating reliance on centralized cloud systems.
xAI's upcoming Grok 4.20 model arrives next week with stronger coding abilities and better front-end design performance compared to its predecessor.
A new survey maps out how AI is shifting from prediction-focused models to autonomous agents that can plan, experiment, and learn from results—bridging reasoning models with robotics and simulation.
Nanbeige LLM Lab has released a compact 3-billion-parameter AI model that punches above its weight class, beating larger Qwen models in reasoning, coding, and agent benchmarks while handling context windows up to 256k tokens.
Major U.S. tech companies are ramping up investments in AI data centers, with spending projected to hit $700 billion in 2026 as massive computing campuses take shape across the country.
OpenAI's GPT-5.2 has uncovered a previously unknown gluon interaction that physicists later confirmed, marking a significant breakthrough in theoretical physics research.
ByteDance just dropped Doubao Large Model 2.0 Pro with Expert Mode, bringing enhanced multimodal and reasoning capabilities to mobile, desktop, and web users.
0G Labs just dropped a toolkit that lets AI coding assistants like Claude and Cursor build directly on its platform, making blockchain development way more accessible through natural language commands.
A compact 3-billion parameter AI model demonstrates performance that rivals significantly larger language models across coding, reasoning, and search benchmarks, signaling a shift toward efficiency in AI development.
Figure showcased its third-generation humanoid robot featuring a seventh-generation robotic hand after years of development. The company claims the design brings them closer to replicating human hand capabilities.
Cline unveiled CLI 2.0, a fully open-source AI coding agent built to run entirely in the terminal, featuring parallel agent execution and native CI/CD automation capabilities.
We use cookies to improve your experience on our site and to show you relevant advertising. To find our more, read our privacy policy and cookie policy
Usman Salis
Eseandre Mordi
Saad Ullah
Marina Lyubimova
Peter Smith
Victoria Bazir
