Muse Spark Open: Meta's 816B Open Model Runs Like a 42B One
Muse Spark Open is Meta's open-weight agentic coding model: 816B parameters, 42B active, 1M context. What it takes to run it, and who should try it first.
Nemotron 3.5 Super VL: Nvidia's Open Model for Always-On Agents
Nemotron 3.5 Super VL is Nvidia's open model for always-on enterprise agents: 120B parameters, 12B active, open data. What it means for self-hosted agents.
ROCm 10.0: AMD's Open AI Stack Now Reaches 90% of H100 Performance
AMD's ROCm 10.0 ships with a new CLI, agentic AI integrations, and an autonomous optimizer. The MI355X now hits 90-95% of H100 throughput at 15-40% lower hardware cost.
Deepseek's 160,000-Chip Huawei Cluster: China's AI Infrastructure Gap, Quantified
Deepseek is building the largest known Huawei compute cluster — but the plan reveals as much about China's chip bottlenecks as its ambitions.
Anthropic's Model Hardware Standard: MCP Arrives in the Physical World
Anthropic's MHS extends MCP to lab hardware, cutting device integration from weeks to hours. Early Carnegie Mellon tests hit 99.3% success — with one significant caveat.
Nvidia's $12.9B Hugging Face Acquisition: Open Source Becomes the Strategic Choke Point
Nvidia is paying 80x revenue to acquire Hugging Face as OpenAI and Anthropic build custom chips — making open-source developer mindshare the most valuable asset left in AI infrastructure.
Cerebras CS-4: 30x Faster Than Nvidia GPUs — and Why That Claim Needs Context
Cerebras' CS-4 uses the same WSE-3 wafer-scale chip as the CS-3 but stacks three wafers per rack instead of two, delivering 4,400 tokens per second per user. Here's what the benchmark means and what it doesn't.
Seven of Eight Agentic AI Pipeline Stages Run on CPUs — and Supply Is Running Out
Agentic pipelines run 7 of 8 stages on CPUs, not GPUs — driving a server CPU shortage that Intel and AMD didn't anticipate until this quarter.
Mojo Is Now Open Source — What It Means for GPU Programming
Mojo's compiler and toolchain went Apache 2.0 on August 18, fulfilling a promise made since May 2023 and opening GPU-optimized Python-syntax programming to community development.
Nvidia's Nemotron 4 Aims for One Trillion Parameters — Chinese Labs Already Passed That
Nvidia is developing a 1-trillion-parameter open-weight model targeting a benchmark that Kimi K3 (2.8T) and DeepSeek V4 Pro (1.6T) have already surpassed. The race is interesting for a different reason.
Anthropic's $5B AMD Deal: Buying 2 Gigawatts of Insurance
Anthropic will deploy up to 2 gigawatts of AMD Instinct MI450 GPUs in a deal worth up to $5 billion — a bet on supplier diversification, not just raw compute.
OpenAI's $230 Codex Micro: When AI Agents Get a Physical Control Surface
OpenAI's first hardware product is a macropad that puts agent status on RGB keys and agent actions on a joystick. Here's what it signals about how developers will work with coding agents.
Bonsai 27B: A 5.9GB LLM That Keeps 95% of Its Reasoning on Your iPhone
PrismML's Bonsai 27B compresses Alibaba's Qwen3.6-27B to 1–2 bits via ternary quantization — 5.9GB for the standard model, 3.9GB for the iPhone variant. Math and coding benchmarks are virtually unaffected.
OpenAI's First Hardware Is a Screenless Speaker Designed to Feel Alive
OpenAI's first hardware product is a portable, screenless AI companion with moving parts and cameras — targeting the home with GPT-Live and a 'Her'-inspired vision of ambient AI.
Nvidia RTX Spark: The Chip That Finally Makes Local AI Agents Practical on Windows
Nvidia's RTX Spark delivers 1 petaflop of AI compute and 128 GB unified memory in a laptop chip — with a security layer built for agent isolation. Here's what it unlocks for on-device AI workflows.
OpenAI Starts With Infrastructure Robots, But the Target Is a Personal Robot for Everyone
OpenAI's physical AI push starts with warehouse and data center robots — not because that's the goal, but because it's the fastest path to the training data needed for general-purpose robotics at consumer scale.
Microsoft and Nvidia's AI PCs: Running Actual Agents Instead of Copilot
Microsoft's second attempt at AI PCs drops the Copilot marketing wrapper and goes deeper — Nvidia chips as main processor, OpenClaw agent framework running locally on Windows. Computex and Build 2026 are the reveal.
Google's Coral Board Runs Gemma 3 Locally — On a Single-Board Computer with 1 TOPS
A RISC-V NPU, 2GB RAM, and 1 TOPS of compute. Google's new Coral board runs Gemma 3 270M entirely offline — and it's targeting headphones, AR glasses, and smartwatches.
Why LLM Scaling Laws Work: MIT's Superposition Explanation
MIT traced LLM scaling to superposition: models pack more concepts than dimensions allow. What this means for AI architecture and interpretability.
Ray-Ban Meta + OpenClaw: Always-On AI Glasses Cut Task Time by Up to 37%, Study Finds
A two-part study from University of Colorado and Gwangju researchers found always-on AI via Ray-Ban Meta smart glasses reduced task time 13–37% and mental demand 7–46% compared to smartphone-based AI assistance.
AI Agents Are Draining the Cloud: Inside the 2026 Compute Crisis
Anthropic's API uptime is below 99%, OpenAI is shutting down Sora to free compute, and GPU prices are up 48%. Agent-driven demand is colliding with infrastructure that wasn't built for it.
Apple Smart Glasses N50: A Sensor, Not a Screen
Apple's display-free smart glasses (codename N50) use computer vision as input for AI — not AR overlays as output. Here's why the architecture matters for agentic AI and how it differs from Meta Ray-Bans.
Nvidia's $26B Open-Source Bet That Could Reshape Agentic AI
Nvidia is spending $26 billion on open-weight AI models. Here's what that shift means for agentic AI infrastructure and the developers who depend on it.