Models & Labs
Hugging Face Launches $399 Open Source Robot, Microduck
TechCrunch AIhigh confidence
Why it matters
- →The Microduck offers an affordable entry point for developers to experiment with physical AI and reinforcement learning.
- →Open-source design allows for greater transparency and customization, appealing to privacy-conscious users.
- →Hugging Face's move into hardware signifies a broader trend of democratizing AI technology beyond software.
Hugging Face has unveiled the Microduck, a $399 open-source robot that aims to make physical AI more accessible. The 25-centimeter-tall robot can perform various tasks, including picking up objects and roller skating, and is equipped with a camera, lidar sensors, and IMUs. Developers can train and deploy behaviors using the provided SDK and simulation tools available on GitHub. This launch follows Hugging Face's acquisition of Pollen Robotics and aligns with their strategy to offer affordable AI hardware.
Read originalMore from TechCrunch AI
© TechCrunch AIMarket & Regulationother
Tech Giants Urge Action Against AI Cyber Threats
In a pivotal move, over a hundred tech companies, including OpenAI, Anthropic, Google, and Microsoft, have come together to call for a unified approach to combat AI-related cyber threats. This collective appeal highlights the increasing sophistication of AI-enabled cyber attacks as AI models continue to evolve. The letter stresses the necessity for innovative cyber defense strategies and urges governments at all levels to work together on security initiatives. This action reflects the complex position of AI companies, who are advancing AI capabilities while also addressing the risks these advancements pose.
TechCrunch AI
© TechCrunch AIGeneral AIproductivity
Google AI Mode Enhances Travel Booking Features
Google's AI Mode is stepping up as a travel assistant, now capable of tracking flight prices and booking hotels directly through its conversational interface. This update transforms AI Mode from a simple search tool into a more interactive travel planner, allowing users to manage their itineraries and receive notifications on price changes. With integration from over 300 airlines and major hotel chains, users can now book trips using points or miles, making travel planning more seamless. This shift positions Google as a significant player in the AI-driven travel booking space, offering a more comprehensive service to users worldwide.
TechCrunch AI
© TechCrunch AIMarket & Regulationbusiness
Google Sets New Memory Standards for Android Apps
Google is responding to a global memory chip shortage by imposing new requirements on Android app developers to optimize memory usage. This move aims to ensure that apps remain efficient and functional even as hardware supply constraints impact device memory availability. Developers are given until February 2027 to comply with these new standards, which include dynamic memory usage and code optimization. By introducing tools to help developers monitor and adjust their apps, Google is proactively addressing potential performance issues, particularly for low-end devices where memory is a critical factor.
TechCrunch AI
More in Models & Labs
Models & Labsmodels
vLLM v0.28.0 Release: Major Performance Enhancements
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.
vLLM Releases
Models & Labsmodels
llama.cpp b10657 Release Expands Platform Support
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.
llama.cpp Releases
Models & Labsmodels
llama.cpp b10658 release adds DFlash2 support
The b10658 release of llama.cpp marks a significant enhancement with the addition of DFlash2, which boosts local convolution and candidate selection capabilities. This update, with contributions from Claude Opus 5, focuses on optimizing costs and refining the code structure for better performance and maintainability. It also resolves several bugs and formatting issues, ensuring a more stable runtime. These improvements make llama.cpp more robust and efficient, catering to developers across various platforms. The release continues to solidify llama.cpp's position as a versatile tool for AI development.
llama.cpp Releases