Intel Crescent Island Bets 480GB of LPDDR5X Against Nvidia's HBM Monopoly
Intel's Xe3P inference accelerator skips HBM entirely, packing more memory than any Nvidia or AMD data center GPU shipping in 2026.
9 min readAnIntent Editorial
AI Infrastructure
Buying guides, explainers, and news covering ai infrastructure
Qualcomm Buys Modular for $3.9B to Crack Nvidia's CUDA Lock-In
Qualcomm's all-stock deal for Chris Lattner's Modular targets the software moat that has protected Nvidia's AI dominance more than any GPU spec.
10 min readAnIntent Editorial
Alphabet Raises $84.75 Billion in Largest US Equity Offering to Fund AI Compute
Alphabet's $84.75 billion equity raise includes $10 billion from Berkshire Hathaway, but roughly $30 billion is earmarked for taxes, not GPUs.
8 min readAnIntent Editorial
OpinionAI Infrastructure
Why Microsoft's Maia 200 Is the Quiet Threat to Nvidia's Inference Lock-In
Maia 200 ships with 216GB HBM3e, native FP4, and a potential Anthropic deal that would crack Nvidia's grip on frontier-model inference.
8 min readAnIntent Editorial
Nvidia Posts $81.6B Q1 FY2027 Revenue, Guides $91B With Zero China Sales
Record quarter, a 25× dividend hike, and an $80 billion buyback land alongside a China Data Center number that has collapsed to zero.
8 min readAnIntent Editorial
ExplainerAI Infrastructure
How Cerebras' Wafer-Scale Engine Works and Why It Rattles Nvidia
The WSE-3 is 57 times larger than an H100 and serves AI tokens from on-chip SRAM at 21 PB/s. Here's what that actually changes.
10 min readAnIntent Editorial
TutorialAI Infrastructure
How to Run Low-Latency AI Inference on the Cerebras CS-3 API
The Cerebras inference API runs Llama 3.3 70B at 2,314 tokens per second, but only if you configure streaming, model choice, and TCP warming correctly.
9 min readAnIntent Editorial
ExplainerAI Infrastructure
MRC Protocol: How OpenAI's Ethernet Rewrite Scales to 100,000 GPUs
OpenAI, Microsoft, and NVIDIA quietly rebuilt Ethernet for frontier AI training. Here's how the Multipath Reliable Connection protocol actually works.
9 min readAnIntent Editorial
Anthropic Leases All of SpaceX's Colossus 1 to Ease Claude Capacity Crunch
Anthropic just took over SpaceX's entire Colossus 1 data center, adding 220,000 GPUs to fix Claude's reliability problems and double Claude Code limits.
9 min readAnIntent Editorial
ExplainerAI Infrastructure
How Panthalassa's Wave-Powered Ocean Nodes Run AI Compute at Sea
Panthalassa's floating nodes generate their own electricity from waves, run AI inference onboard, and beam tokens to shore by satellite. Here's the actual mecha
9 min readAnIntent Editorial