Models & Labs
Apple's AI Chips Emerge from Self-Driving Car Effort
The Verge AIhigh confidence
Why it matters
- →Apple's AI chips originated from its self-driving car project, showing how failed initiatives can lead to breakthroughs in other areas.
- →The Neural Engine has become a key component in Apple's devices, enhancing on-device AI capabilities and privacy.
- →The upcoming M7 Ultra chip could significantly boost AI processing power, supporting new server products and advanced applications.
Apple's self-driving car program may not have succeeded, but it catalyzed the development of powerful AI chips. The need for advanced on-device AI processing led to the creation of the Neural Engine, first seen in the iPhone X. This innovation has become central to Apple's hardware, supporting features like FaceID and augmented reality. As Apple accelerates the development of its M7 Ultra chip, expected to support up to 1.5TB of RAM, the company continues to leverage its AI hardware prowess, despite lagging in AI software.
Read originalMore from The Verge AI
© The Verge AIMarket & Regulationbusiness
Community Pushback Against AI Data Centers Grows
The expansion of AI data centers is facing significant resistance from communities across the US and beyond. Concerns over rising energy costs, environmental impact, and local disruptions are prompting residents to protest and block new projects. This grassroots opposition has already led to the cancellation or delay of numerous data center plans, highlighting the growing tension between technological advancement and community welfare. As political battles unfold, with some lawmakers pushing for stricter regulations, the future of AI data center expansion remains uncertain. This resistance marks a pivotal moment in balancing technological growth with community and environmental considerations.
The Verge AI
© The Verge AIMarket & Regulationimage
Meta Disables Instagram AI Deepfake Feature
Meta has quickly removed a controversial Instagram feature that allowed users to create AI-generated images using content from public accounts. This feature, part of Meta's Muse Image AI model, faced heavy criticism for enabling potential misuse without the consent of account owners. Critics, including the National Center on Sexual Exploitation, pointed out risks such as sextortion and privacy violations. Initially, Meta offered an opt-out option hidden in settings, but the backlash led to the feature's complete removal. This decision highlights the ongoing challenge for tech companies to innovate while respecting user privacy and safety.
The Verge AI
© The Verge AIMarket & Regulationbusiness
Apple sues OpenAI over alleged trade secret theft
Apple has taken legal action against OpenAI, accusing the company of illicitly acquiring trade secrets to enhance its hardware development. The lawsuit alleges that former Apple employees, now working at OpenAI, transferred confidential information about Apple's unreleased products and processes. This legal conflict reveals the intense competition among tech giants as they venture into AI hardware. If the allegations hold true, OpenAI's hardware plans and its partnerships could face significant challenges. This case brings attention to the vital role of intellectual property protection in the tech industry.
The Verge AI
More in Models & Labs
Models & Labsmodels
vLLM v0.25.0 Release Enhances Model Execution
The vLLM v0.25.0 release marks a significant step forward in model execution and performance. With Model Runner V2 now the default for dense models, users benefit from enhanced support for real-time embeddings and dynamic speculative decoding. The removal of PagedAttention and improvements to the Transformers backend, including FP8 MoE support, streamline operations and boost speed. New models like LLaVA-OneVision-2 and Unlimited OCR expand the model zoo, offering more options for developers. This release solidifies vLLM's position as a robust platform for AI model deployment, with improved efficiency and expanded capabilities.
vLLM Releases
Models & Labsmodels
llama.cpp b9974 Release Fixes CUDA Memory Crash
The latest b9974 release of llama.cpp addresses a critical issue for CUDA users by preventing crashes when querying memory on devices with no free memory. Previously, attempting to check memory availability could lead to a fatal crash if the device was out of memory. The update now assigns zero total/free memory to such devices, ensuring the fit algorithm doesn't attempt to use them, thus avoiding crashes. This change enhances stability for CUDA-enabled builds, especially when users specify '-dev none'. While the update doesn't introduce new features, it significantly improves reliability for developers working with CUDA devices.
llama.cpp Releases
Models & Labsmodels
Llama.cpp b9977 release fixes image block conversion
The latest b9977 release of llama.cpp addresses a critical issue in the conversion process between Anthropic and OpenAI formats, where image blocks in tool results were being dropped. This fix ensures that multimodal tool outputs, such as those returning images, are correctly processed and received by the model. By converting image blocks into OpenAI's multimodal content parts, the update maintains backward compatibility with plain-text results. This release is a technical refinement that enhances the robustness of multimodal AI applications, ensuring seamless integration across different platforms.
llama.cpp Releases