Category
Business Insider reports that Hugging Face is exploring a sale at a valuation of more than $13 billion. Separating ownership of public models from operational control of the platform reveals what's really being priced: the neutrality of AI's distribution layer.
NVIDIA's Groq 3 LPX, now in mass production, splits AI inference work between GPUs and LPUs to reach 3,431 output tokens per second. Here's what that number means, where the benchmark falls short, and what's behind the $17 billion deal.
NVIDIA is reportedly discussing a follow-on investment in Perplexity at a valuation exceeding $30 billion. The talks come as Perplexity's agentic business grows and its ties with NVIDIA span GPUs, CPUs, and model development.
Geely has begun real-world testing of its solid-state batteries and plans to scale up to a 1,000-vehicle trial by 2027. Whether the measurement basis for its 500Wh/kg claim and manufacturing costs hold up will determine the fate of its 2030 mass-production plan.
Cerebras' CS-4 keeps the same 5nm wafer but redesigns power delivery, cooling, and I/O to double peak compute per wafer. Here's what the performance claims actually cover and how production-ready the system really is.
Xiaomi's second-generation SoC, the XRING O3, stays on 3nm while expanding to 10 high-performance CPU cores, a 16-core GPU, LPDDR6, and a 200TOPS NPU. This article separates official figures from measurements on a development board to examine what the 113.8GB/s memory bandwidth and 20W-class peak power actually mean.
d-Matrix's Raptor accelerator measured roughly 105 TB/s of card-level bandwidth using 3D-DRAM in real silicon. But its inference speedups come from architectural modeling, not third-party benchmarks, and production yield data remains unpublished.
WVU's A34 facility achieved over 95% purity rare earth oxides from acid mine drainage. While the process was demonstrated on real wastewater, the 4-ton annual figure represents capacity, not verified output, leaving recovery rates and profitability still unconfirmed.
The anonymous model Ox Alpha has different data retention terms depending on the access route, and even within OpenRouter, the model description and Stealth Program terms conflict. Here's why companies should evaluate performance and data contracts separately.
Cursor's Git infrastructure treats an S3-based write-ahead log as the source of truth, with local NVMe serving as a regenerable cache. We examine the design's strengths and open questions as it consolidates write ordering into S3 CAS and decouples replica count from push latency.
