témata
AI modely
Vydání modelů, jejich schopnosti, hodnocení a praktické změny za benchmarky.
Nejnovější články
Časová osa
- Dva režimy AGI: DeepMind chrlí produkty, Anthropic vyjednává s Bílým domem
- Anthropic offline: jak Bílý dům začal psát pravidla pro AI za pochodu
- The week the inference layer took over.
- Export controls meet open weights
- The quiet rewiring of the developer stack
- The weekend the context windows cracked open
Opakující se entity
Hlavní citované zdroje
- Investing in multi-agent AI safety research · 2
- Claude Opus 4.7 with 1M context · 2
- Introducing computer use in Gemini 3.5 Flash · 1
- The Trump Administration Is Lifting Its Export Controls on Anthropic’s Mythos and Fable AI Models · 1
- Anthropic Added a New Security Measure to Get Back Into the Trump Administration’s Good Graces · 1
- Claude Helped a Hacker Find a Way to Issue Tickets to Almost Every US Music Festival · 1