NVIDIA has launched the Jetson Orin Nano 2, an edge robotics computer designed to enhance AI capabilities in drones, robots, and vision systems. The board offers 78 trillion operations per second of AI compute and features an eight-core Arm CPU with 8GB of memory. It delivers twice the inference performance of its predecessor while using 40% less power. NVIDIA aims to enable real-time AI processing at the edge, allowing for advanced applications in robotics and autonomous systems. Early adopters include companies like Wing and Matic Robots, exploring its potential for drone delivery and home cleaning robots.
Read originalNvidia is strategically investing in AI labs, which in turn purchase its chips, creating a cycle that could account for a quarter of its business next year. By partnering with major investment firms to raise over $500 billion, Nvidia is ensuring these labs have the resources to build data centers filled with its technology. This approach, while labeled as circular financing by some, is seen by Nvidia as a way to meet the high demand for computing power that young AI companies face. The move not only boosts Nvidia's revenue but also strengthens its position in the AI hardware market, despite potential risks if demand were to falter.
The v0.28.0 release of vLLM introduces substantial improvements in performance and functionality, particularly for the Kimi-K3 model. With the addition of Decode Context Parallel support and fused FlashKDA decode kernels, the update significantly enhances processing speed and efficiency. DeepSeek V4 now includes sparse MLA support and advances in speculative decoding, offering better execution on both NVIDIA and AMD hardware. These updates make vLLM more robust and adaptable, providing developers with enhanced tools for deploying and executing models on a broader range of hardware configurations.
The b10657 release of llama.cpp brings new OpenCL binary kernels, enhancing performance and compatibility across a wide range of systems. This update includes specific improvements for Apple Silicon, with KleidiAI support, and Vulkan on Ubuntu, making it more accessible for developers using these platforms. While no new model architectures are introduced, the release focuses on strengthening llama.cpp's capabilities as an inference runtime, particularly for those not using NVIDIA hardware. With ROCm 7.14 support on Ubuntu and CUDA 12 and 13 DLLs for Windows, llama.cpp continues to evolve as a versatile tool for AI model deployment. This release underscores the commitment to broadening hardware compatibility and optimizing performance across different environments.
Gatik's $200 million Series D funding round marks a significant step in scaling its AI-powered autonomous freight operations across North America. With backing from major investors like Qatar Investment Authority and Koch Disruptive Technologies, Gatik plans to expand its fleet of driverless trucks, which have already completed 85,000 fully autonomous deliveries. The company is focusing on middle-mile logistics, moving goods between distribution centers and retail locations, and aims to increase its fleet to thousands of trucks. This funding will enable Gatik to enhance its technology and infrastructure, potentially transforming regional freight networks with dynamic route orchestration.