Data
Display
Estimates are derived by T2C from disclosed figures. Switch them off to see only what companies have published.
Focus mode Supporting context hidden. Figures, their basis, confidence and sources are unchanged — nothing that qualifies a number is hidden.

AI news

What is moving across the AI supply chain

The latest stories from every stage — chips to power to the models themselves — and which neoclouds and AI-infrastructure operators each one touches through a disclosed supplier or customer link.

Latest stories

Loading…
  1. Loading the latest stories…

Headlines are third-party and shown as published. Stock-tip and valuation pieces are hidden as market commentary — one tap shows them. “Touches” lists the operators a named company is linked to by a filing — exposure, not a prediction, and never investment advice.

The models the compute is for

Every model, with prices →
Frontier reasoning

OpenAI

GPT-6 Astra

“Our most capable model for the most demanding work.”

What people build with it T2C summary

  • Hard analysis and research
  • Complex multi-step agents
OpenAI ↗
Frontier reasoning

Anthropic

Claude Fable 5.1

“For demanding reasoning and long-horizon agentic work”

What people build with it T2C summary

  • Long-running autonomous agents
  • Deep research and analysis
Anthropic ↗
Frontier reasoning

Google

Gemini 4 Argon

“Our flagship reasoning model with a unified 1M context window and long decodes up to 1M tokens.”

What people build with it T2C summary

  • Reasoning over very long documents
  • Large codebases
Google ↗
Coding and agents

Meta

Muse Wonder

“Frontier coding, long-running agents, and advanced computer use.”

What people build with it T2C summary

  • Computer-use agents
  • Coding
Meta ↗
Frontier reasoning

Microsoft

MAI-Thinking-1

“Competitive reasoning and top SWE-Bench Pro results at a mid-weight price.”

What people build with it T2C summary

  • Reasoning at lower cost
  • Software fixes
Microsoft ↗
Frontier reasoning

xAI

Grok 4.7

“For everything else, including code, use Grok 4.7. It is the most capable model we've built.”

What people build with it T2C summary

  • Chat
  • Coding
xAI ↗