Explainers
Evergreen explainers on AI models, labs, chips, data centers, benchmarks, robotics and events.

DeepSeek models: V-series and R-series by date
DeepSeek's model releases by date, from V3 and R1 to V4, drawn from its own transparency page and changelog.

ARC-AGI explained: ARC-AGI-3 scoring, prizes and scores
How ARC-AGI-3 scores agents against human action counts, what ARC Prize 2026 pays, and where frontier models stand.

When does OpenAI go public? What OpenAI has said
OpenAI has set no date to go public; it confidentially submitted a draft S-1 to the SEC on June 8, 2026.

Humanoid robot battery life: runtime and charging by robot
How long humanoid robots run per charge, from 1 to 5 hours, and how charging and swaps set their working day.

Gemini models list: every Gemini API model by date
Every Gemini API model by date, from 2.0 Flash to Gemini 4 Hydrogen, with IDs, status and the December 1 changes.

NeurIPS 2026: Sydney, Atlanta and Paris dates, venues and fees
NeurIPS 2026 in Sydney, Atlanta and Paris: dates, venues and fees by site, with who can still register.

Frontier AI safety framework comparison: RSP vs Preparedness vs FSF
How five labs' frontier AI safety frameworks compare on thresholds, who decides and where new models landed.

AMD Taalas acquisition: the $2.9 billion price and what Taalas builds
AMD's 10-Q puts the Taalas deal at about $2.9 billion, up to $1.4 billion contingent, plus the HC1 chip's specs.

Anthropic IPO: S-1, pricing and Nasdaq listing as ANT
Anthropic's IPO was declared effective and priced on Nov 18, 2026; the Nasdaq listing date for ANT is Nov 19.

OpenAI models list: every GPT and o-series model by date
Every OpenAI GPT and o-series model from GPT-4 to GPT-6.1 Terra, with release dates and API status as of November 2026.

METR time horizons: what the AI agent benchmark measures
How METR estimates AI task-completion time horizons, what the 50% threshold means and where the benchmark's limits lie.

Where AI data centers are being built: selected projects
Selected announced AI data-center sites, with their regions, capacity definitions and dated development milestones.

AI inference vs training compute: what's the difference
Training builds a model and inference runs it, and the two stress AI hardware in different ways.

Microsoft Fairwater AI data centers explained
Fairwater is Microsoft's design for AI datacenters that run as one GPU cluster per site, linked by a dedicated AI WAN.

Meta Hyperion and Prometheus: Meta's AI data centers explained
Hyperion is Meta's AI cluster in Richland Parish, Louisiana, now planned at 5 GW; Prometheus is in New Albany, Ohio.

xAI Colossus: the Memphis supercomputer explained
xAI Colossus is the AI training supercomputer xAI built in a former Electrolux plant in Memphis.

Vision-Language-Action Models: From Instructions to Motion
How VLA models combine robot vision, instructions and sensor state to produce physical actions.

Huawei Ascend: China's AI chips explained
Huawei Ascend AI chips from the 910 series to the 950, 960, 970 and 980, as Huawei described them at Huawei Connect.

Intel Gaudi explained: specs and where it stands
Intel Gaudi explained: Gaudi 3 specs from Intel's own paper, where to get it, and Intel's next AI chip.

HBM suppliers: SK hynix, Samsung and Micron in the HBM race
SK hynix, Samsung and Micron all say they ship HBM4 in volume as of October 2026; here is where each supplier stands.

Claude models list: every Anthropic model by date
Every Claude model since Claude 4 by date, with the current line-up's API IDs, context windows and retirement dates.

TSMC and AI chips: why so many accelerators run through it
TSMC makes and packages the silicon behind Nvidia GPUs, Broadcom XPUs and Intel Gaudi 3: nodes, CoWoS and Arizona.

Nvidia GR00T explained: the Groot robot foundation model, N1 to N1.7
How Nvidia GR00T turns camera images and instructions into robot motion, and what changed from N1 to N1.7.

Broadcom AI chips: custom XPUs and who designs them
Broadcom's AI chips are custom accelerators (XPUs) built for Google, Meta, OpenAI and others.
Each explainer answers a stable question from primary sources: how a chip generation differs, what a benchmark measures, a lab’s model line-up by date or what an AI data center is. Browse them by subject: AI model explainers, AI lab explainers, AI chip explainers, AI data center explainers, AI benchmark explainers, Humanoid robot explainers and AI event explainers.
