Discover quality AI tools
Bring together the world’s leading artificial intelligence tools to help improve work efficiency
Articles published by RogerMartinez
X Square Robot open-sourced Wall-OSS-0.5, a Vision-Language-Action model for embodied intelligence. It achieved zero-shot deployment on real robots across 17 challenging tasks without fine-tuning, scoring up to 82 on unseen flexible object tasks. Compared to industry benchmark π0.5, it led by 17.5 points on average and showed near order-of-magnitude improvement in precision operations. Key innovations include gradient bridging, visual alignment tokenizer, action space supervision, and DMuon distributed optimization. The model weights, training code, and dataset interfaces are fully open-sourced.
X Square Robot open-sourced Wall-OSS-0.5, a Vision-Language-Action model for embodied intelligence. It achieved zero-shot deployment on real robots across 17 challenging tasks without fine-tuning, scoring up to 82 on unseen flexible object tasks. Compared to industry benchmark π0.5, it led by 17.5 points on average and showed near order-of-magnitude improvement in precision operations. Key innovations include gradient bridging, visual alignment tokenizer, action space supervision, and DMuon distributed optimization. The model weights, training code, and dataset interfaces are fully open-sourced.
Anthropic signs a $1.25B/month deal to lease 300MW of computing power from xAI's Tennessee data center until 2029, a move expected to generate over $40B for xAI. This 'neocloud' model allows AI firms to monetize excess capacity, providing Anthropic critical resources while helping xAI utilize idle infrastructure.
Anthropic signs a $1.25B/month deal to lease 300MW of computing power from xAI's Tennessee data center until 2029, a move expected to generate over $40B for xAI. This 'neocloud' model allows AI firms to monetize excess capacity, providing Anthropic critical resources while helping xAI utilize idle infrastructure.
Former Microsoft VP Eric Boyd joins Anthropic as head of infrastructure. He will lead the build-out of large-scale AI foundational systems, leveraging his experience from commercializing Azure Machine Learning and managing platforms for models like Claude. This move comes as Anthropic, facing explosive demand and service stability challenges, transitions to a full-scale tech provider. The company plans a $50 billion investment in U.S. AI data centers and deeper collaboration with Google on TPUs. The AI industry's competition is increasingly defined by infrastructure scalability and cost-efficient inference capabilities.
Former Microsoft VP Eric Boyd joins Anthropic as head of infrastructure. He will lead the build-out of large-scale AI foundational systems, leveraging his experience from commercializing Azure Machine Learning and managing platforms for models like Claude. This move comes as Anthropic, facing explosive demand and service stability challenges, transitions to a full-scale tech provider. The company plans a $50 billion investment in U.S. AI data centers and deeper collaboration with Google on TPUs. The AI industry's competition is increasingly defined by infrastructure scalability and cost-efficient inference capabilities.
Meituan's LongCat team introduced LongCat-Flash-Lite, a language model using a novel embedding expansion approach instead of scaling experts. With 68.5B total parameters, it activates only 2.9B-4.5B per inference via an efficient N-gram embedding layer. System optimizations enable speeds of 500-700 tokens/s. It leads in agent and coding benchmarks like SWE-Bench (54.4%) and scores 85.52 on MMLU. The model is now fully open-sourced.
Meituan's LongCat team introduced LongCat-Flash-Lite, a language model using a novel embedding expansion approach instead of scaling experts. With 68.5B total parameters, it activates only 2.9B-4.5B per inference via an efficient N-gram embedding layer. System optimizations enable speeds of 500-700 tokens/s. It leads in agent and coding benchmarks like SWE-Bench (54.4%) and scores 85.52 on MMLU. The model is now fully open-sourced.
Chinese apparel brand Hongdu partners with China Electronics Technology Group to launch an AI wearable robot. Featuring lightweight exoskeleton technology with multi-modal sensing and lower limb intent recognition, it assists movement for activities like walking and hiking. This marks a key step in Hongdu s AI plus apparel strategy, extending comfort from clothing to mobility, with future plans for AI retail and robot services.
Chinese apparel brand Hongdu partners with China Electronics Technology Group to launch an AI wearable robot. Featuring lightweight exoskeleton technology with multi-modal sensing and lower limb intent recognition, it assists movement for activities like walking and hiking. This marks a key step in Hongdu s AI plus apparel strategy, extending comfort from clothing to mobility, with future plans for AI retail and robot services.
Global fintech venture funding rose 27% to $51.8 billion in 2025, reaching a multi-quarter high driven by larger late-stage deals. However, deal count fell 23% to 3,457, indicating concentrated capital. Major rounds included Polymarket, Binance, and Kalshi. Funding now exceeds pre-pandemic levels but remains below 2021's peak.
Global fintech venture funding rose 27% to $51.8 billion in 2025, reaching a multi-quarter high driven by larger late-stage deals. However, deal count fell 23% to 3,457, indicating concentrated capital. Major rounds included Polymarket, Binance, and Kalshi. Funding now exceeds pre-pandemic levels but remains below 2021's peak.
OpenAI launches GPT-5 with competitive pricing at $1.25 per million input tokens and $10 per million output tokens, undercutting Anthropic’s Claude Opus 4.1 and matching Google’s Gemini 2.5 Pro, potentially sparking an AI model price war.
OpenAI launches GPT-5 with competitive pricing at $1.25 per million input tokens and $10 per million output tokens, undercutting Anthropic’s Claude Opus 4.1 and matching Google’s Gemini 2.5 Pro, potentially sparking an AI model price war.
