China Activates 1,243-Mile AI Computing Network: 98% Efficiency at Continental Scale
China's Future Network Test Facility spans 1,243 miles, achieving 98% single-datacenter efficiency. Xi touts AI breakthroughs. $70B investment wave coming.
Blog
Insights on GPU infrastructure, AI, and data centers.
Thinking Without Tokens: How Latent Reasoning Makes a 3.5B Model Perform Like 50B
A new architecture lets AI models reason in latent space instead of generating tokens. The recurrent depth approach scales test-time compute without chain-of-thought.
LTX-2: The First Open-Source Model That Generates Synchronized Video and Audio
Lightricks releases LTX-2 with 14B video + 5B audio parameters. Native 4K at 50fps with lip sync, foley, and ambient sound. Fully open weights.
Recursive Language Models: Teaching AI to Manage Its Own Context
MIT's RLM architecture lets models delegate context to sub-LLMs and Python scripts. 100x context extension with 2-3x token efficiency. Prime Intellect predicts the paradigm of 2026.
MiroThinker: The Third Scaling Dimension for AI Agents
MiroThinker introduces interaction scaling—training agents to handle 600 tool calls per task. 81.9% on GAIA benchmark. A new dimension beyond model size and context.
Microsoft's $60B Neocloud Bet: Buying Time in the AI Capacity Crisis
Microsoft commits $60B+ to neocloud providers including $23B to Nscale for 200K GB300 GPUs. Azure capacity crunch extends into 2026. Neoclouds reshape AI infrastructure.
UPS and Power Distribution for AI: Designing Resilient 2N+1 Infrastructure
A 47-second power interruption at Meta's data center caused $65 million in losses when 10,000 GPUs performing distributed training lost synchronization, corrupting three weeks of model progress.
Disaggregated Computing for AI: Composable Infrastructure Architecture
CXL memory pooling achieves 3.8x speedup compared to 200G RDMA and 6.5x speedup compared to 100G RDMA when sharing memory across GPU servers running large language model inference. The
Trump Opens H200 Exports to China with 25% Surcharge
Trump reverses Biden export restrictions, allowing NVIDIA H200 sales to China with 25% surcharge. Blackwell GPUs remain restricted. Tariff-based controls replace bans.
Feature Stores and MLOps Databases: Infrastructure for Production ML
Uber's Michelangelo feature store processing 10 trillion feature computations daily, Airbnb's Zipline serving features with sub-10ms latency to millions of models, and DoorDash's Fabricator reducing
Retrofitting Legacy Data Centers for AI: Liquid Cooling Integration Guide
A 15-year-old data center designed for 5kW racks now faces demands for 40kW GPU clusters, creating an infrastructure crisis that forces organizations to choose between $50 million new facility
DeepSeek V3.2 Beats GPT-5 on Elite Benchmarks: What China's AI Surge Means for Infrastructure
DeepSeek V3.2-Speciale achieves 96% on AIME, gold at IMO, and top-10 at IOI—matching U.S. frontier models despite export restrictions.